17
Research Article Leveled Design of Cryptography Algorithms Using Cybernetic Methods for Using in Telemedicine Applications Ali Mohammad Norouzzadeh Gil Molk, 1 Mohammad Reza Aref , 2 and Reza Ramazani Khorshiddoust 3 1 Department of Computer Engineering, Islamic Azad University, North Tehran Branch, Tehran, Iran 2 Department of Electrical Engineering, Sharif University of Technology, Tehran, Iran 3 Department of Industrial Engineering & Management Systems, Amirkabir University of Technology, Tehran, Iran Correspondence should be addressed to Mohammad Reza Aref; [email protected] Received 28 May 2021; Revised 31 July 2021; Accepted 12 August 2021; Published 11 September 2021 Academic Editor: Navid Razmjooy Copyright © 2021 Ali Mohammad Norouzzadeh Gil Molk et al. is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. e technology world is developing fast with the developments made in the hardware and software areas. Considering that privacy and security of telemedicine applications are among the main necessities of this industry, as a result, there is a need to use lightweight and practical algorithms to be used in applications in the field of telemedicine, while security have the least negative impact. e distinct and contradicting components in the design and implementation of the cryptography algorithm, to achieve various objectives in medicine-based applications, have made it a complicated system. It is natural that, without identifying the components, indices, and properties of each system component, the hardware and software resources are lost and a proper algorithm cannot be designed. Accordingly, this paper presents a leveled model of cryptography algorithms using the cybernetic method. First, the main objectives and measures in the design of the cryptography algorithms are extracted using the measure reduction methods, and some of the excess and overlapping measures are eliminated. en, three general classes of the cryptography algorithm design and implementation measures, applications of cryptography algorithms, and cryptography implementation techniques are extracted. Since the complexity of the cryptography algorithm design is relatively high, the cybernetic methodology is used to present a supermodel to make the cryptography algorithm design objective. Such design prevents examining unnecessary details and establishes a bidirectional relationship between the main design and implementation process and the support process. is relationship provides the support requirements of the main process by the support process at each step. Finally, the Q-analysis tools are used to analyse the proposed method, and the efficiency results are represented. 1. Introduction Since telemedicine technology relies on data transmission, data security is critical in order to keep information transmission confidential and patients’ privacy, and any potential threat or attack on telemedicine networks such as unauthorized access to data and alteration or destruction of patient data should be considered. In other words, any weakness in any part of the telemedicine network can affect the entire system. erefore, in order to create security in the field of storage and exchange of information in the medical network, enforcement mechanisms using relevant standards should be considered. Accordingly, this study has focused on the surface design of cryptographic algorithms for use in telemedicine. Security of cryptography systems depends on “algorithm power” and “key size,” [1] and the general cryptography levels are divided into three levels, including cryptography algorithms, security protocols, and applica- tions [2, 3]. However, the cryptography algorithms are designed and implemented to achieve goals such as confi- dentiality, authentication, and integrity [4], but various components such as speed, resource consumption, appli- cation type, flexibility, scalability, and reliability should be considered for their design. Design of a cryptography Hindawi Computational Intelligence and Neuroscience Volume 2021, Article ID 3583275, 17 pages https://doi.org/10.1155/2021/3583275

Leveled Design of Cryptography Algorithms Using Cybernetic

  • Upload
    others

  • View
    2

  • Download
    0

Embed Size (px)

Citation preview

Research ArticleLeveled Design of Cryptography Algorithms Using CyberneticMethods for Using in Telemedicine Applications

Ali Mohammad Norouzzadeh Gil Molk1 Mohammad Reza Aref 2

and Reza Ramazani Khorshiddoust3

1Department of Computer Engineering Islamic Azad University North Tehran Branch Tehran Iran2Department of Electrical Engineering Sharif University of Technology Tehran Iran3Department of Industrial Engineering amp Management Systems Amirkabir University of Technology Tehran Iran

Correspondence should be addressed to Mohammad Reza Aref arefsharifedu

Received 28 May 2021 Revised 31 July 2021 Accepted 12 August 2021 Published 11 September 2021

Academic Editor Navid Razmjooy

Copyright copy 2021 Ali Mohammad Norouzzadeh Gil Molk et al is is an open access article distributed under the CreativeCommons Attribution License which permits unrestricted use distribution and reproduction in any medium provided theoriginal work is properly cited

e technology world is developing fast with the developments made in the hardware and software areas Considering that privacyand security of telemedicine applications are among the main necessities of this industry as a result there is a need to uselightweight and practical algorithms to be used in applications in the field of telemedicine while security have the least negativeimpact e distinct and contradicting components in the design and implementation of the cryptography algorithm to achievevarious objectives in medicine-based applications have made it a complicated system It is natural that without identifying thecomponents indices and properties of each system component the hardware and software resources are lost and a properalgorithm cannot be designed Accordingly this paper presents a leveled model of cryptography algorithms using the cyberneticmethod First the main objectives and measures in the design of the cryptography algorithms are extracted using the measurereduction methods and some of the excess and overlapping measures are eliminated en three general classes of thecryptography algorithm design and implementation measures applications of cryptography algorithms and cryptographyimplementation techniques are extracted Since the complexity of the cryptography algorithm design is relatively high thecybernetic methodology is used to present a supermodel to make the cryptography algorithm design objective Such designprevents examining unnecessary details and establishes a bidirectional relationship between the main design and implementationprocess and the support processis relationship provides the support requirements of themain process by the support process ateach step Finally the Q-analysis tools are used to analyse the proposed method and the efficiency results are represented

1 Introduction

Since telemedicine technology relies on data transmissiondata security is critical in order to keep informationtransmission confidential and patientsrsquo privacy and anypotential threat or attack on telemedicine networks such asunauthorized access to data and alteration or destruction ofpatient data should be considered In other words anyweakness in any part of the telemedicine network can affectthe entire systemerefore in order to create security in thefield of storage and exchange of information in the medicalnetwork enforcement mechanisms using relevant standards

should be considered Accordingly this study has focused onthe surface design of cryptographic algorithms for use intelemedicine Security of cryptography systems depends onldquoalgorithm powerrdquo and ldquokey sizerdquo [1] and the generalcryptography levels are divided into three levels includingcryptography algorithms security protocols and applica-tions [2 3] However the cryptography algorithms aredesigned and implemented to achieve goals such as confi-dentiality authentication and integrity [4] but variouscomponents such as speed resource consumption appli-cation type flexibility scalability and reliability should beconsidered for their design Design of a cryptography

HindawiComputational Intelligence and NeuroscienceVolume 2021 Article ID 3583275 17 pageshttpsdoiorg10115520213583275

algorithm should be systematic comprehensive and stagedAll required components of the information security shouldbe considered in an excellence pattern in terms of technicalorganization procedural and humanitarian aspects Iden-tifying new cryptographic challenges such as post-quantumcryptography and its agility mobile applications robustnessof algorithms and the role of implementation methods toachieve the above goals can be implemented in a compre-hensive model [5ndash7] Providing all these requirements si-multaneously in the design of an algorithm is difficult andsometimes impossible If contradictory objectives are con-sidered for formulation of an objectiveobjectives of analgorithm most algorithms might be broken and if attackerhas sufficient time motivation and resources he can trackthe information [8] Accordingly presenting a model tomake the cryptography algorithm design targeted is veryimportant erefore in this study an approach is presentedto design a leveled model of cryptography algorithms usingthe cybernetic method e rest of this paper is structured asfollows In the second section the literatures review ispresented In the third section the cybernetic methodologyis described and a model is specified for design andimplementation of cryptography algorithms based onextracted indices In the fourth section the proposed modelis used to examine the design of the cryptography algorithmusing the cybernetic supermodel Finally the proposedapproach is evaluated using the Q-analysis method

2 Literature Review

In [9] a security approach based on cryptography has beenpresented through examining the security issues in mobiledevices and the available solutions Also it is mentioned thatasymmetric cryptography is not a proper option for securingthe resource-limited infrastructures such as IoT due to highcomplexity of the design and implementation On thecontrary employing symmetric algorithms has other se-curity issues Accordingly it has studied the design ofcryptography algorithms based on position To this end anapproach based on the AES algorithm and position of anefficient cryptography approach has been designed In thisapproach the application diagram is described and the useroperation is studied In the following the operation flows ofthe system are described Finally it was evaluated that se-curity can be increased through employing this approach In[10] the design of stream cipher algorithms has beenstudied It has been mentioned that stream cipher is one ofthe essential branches of symmetric cryptography whichrequires limited hardware resources for execution ere-fore considering the development of the communicationtechnologies the need to these algorithms is increasingAccordingly in the following the design procedures andperformance of various encryption algorithms includingNFSR eStram FCSR and Panama are presented throughdescribing the main requirements of the cryptography al-gorithm design e main purpose of this study is to presenta perspective of stream cipher algorithm design and theirperformance In 2016 NIST published a document calledcipher standard and development instructions [11]

Transparency openness balance accuracy technical meritglobal acceptability usability continuous improvement andinnovation and intellectual property (IIP) are the guidanceprinciples of NIST cipher standards and developmentprocedures Also NIST has started a procedure to requestevaluate and standard of one or multiple public key cipheralgorithms robust against quantum attacks [12] In [11]lifecycle management processes and policies of cipherstandard have been presented where its main principlesinclude identifying and evaluating the needs announcingthe userrsquos intention on a standard or instruction consideringthe requirements and solutions defining a specific programand procedure and design and development of a standardand evaluating and maintaining the standard In [13] ananalytical framework has been presented to hardware andsoftware implementation using cipher programs that verifiesan integrated statistical framework which can implement theclassified algorithms successfully based on a combination ofheterogeneous hardware features and their software appli-cations e model presented in this paper includes six el-ements of goal input activities output outcomes andperformance In [14] software engineering methodologieshave been used to propose an adaptive approach for pre-senting a robust cipher key generation algorithm etechnique used in this method is based on self-checkingprocedures that can detect the system-level errors ere-fore it can be used to check the security keys generated viaemploying random factors ese factors have been pre-sented in the NIST evaluation results In this softwaremethod the values of the random factors are smaller thanthe acceptance values and the key is generated when a validvalue is detected e generated keys are generated throughshift register and SIGBA technique e evaluation resultsindicate the efficiency of the presented approach in gener-ating valid cipher keys In [15 16] security issues of mobiledevices and processing infrastructure including mobilecomputing and edge computing have been studied It hasbeen concluded the importance of cipher algorithms andnecessity of employing new models consider the complexityof these infrastructures Bhowmik et al [17] focused onsecurity issues in telemedicine and introduced a double-tier(nDTCS) encryption system Accordingly this solution hasproposed a modified logistic map and a congruence-basedsecurity model to secure telemedicine medical transactionsTwo keys have been used for the encryption and decryptionprocess intermediate key and session key e results of theevaluation indicate the effectiveness of the solution in orderto secure the information through the proposing method[18] In order to protect telemedicine communications a keyexchange solution is proposed by improving the Dif-fiendashHellman cryptographic algorithm In this method arandomized key generation is used to generate the key eproposed solution is naturally safe and reliable due to the useof the DiffiendashHellman algorithm so there is no need forrecalculations or key reversal e results of the evaluationalso indicate that the proposed method is safe againstguessing key attacks In [19] an intelligent and securedtransmission security solution for heart disease reportsbased on session key-based methods is presented For this

2 Computational Intelligence and Neuroscience

purpose matrix confusion operations are used e inno-vation of this solution is in the process of matrix transferwhich is transmitted in the form of a number of cardiologistsin particular Finally the efficiency of the proposed methodis evaluated with regard to cryptographic engineeringtransparency and strength e results indicate that thismethod provides more security in the medical data trans-mission process Hosseinian et al [20] examined the im-portance of information security needs in telemedicinetechnology in the field of information transmission In thisstudy the data collection tool was a questionnaire that wasdesigned based on the criteria of the Association of Infor-mation Management and Health Care Systems (HMISS) inthe field of telemedicine network security and securitystandards of the American Telemedicine Association isquestionnaire has been calculated separately based on ascore of 1 to 5 Table 1 summarizes the importance of eachsection

3 Cybernetic Supermodel

Considering the complexity of the cipher context in terms ofvarious aspects designing a cipher algorithm should besystematic comprehensive and stage To design cipher al-gorithms different technologies including mathematicsphysics biometric biology and social engineering are used[21 22] Also concepts and basic sciences such as theory ofnumbers Boolean functions [23] and random functions[24 25] are very essential Depending of the application ofcipher algorithms various technical and nontechnical re-quirements should be considered for their design Detectingnew cryptography challenges such as postquantum cipherand its agility [26] and mobile applications [6 7] making analgorithm robust and the role of implementation methodsto achieve the above goals are the issues that should beconsidered in a comprehensive model Amidst consideringthe large number of components in the design of cipheralgorithms and their relationship and impact on each otherthe design and implementation of these algorithms hasbecome complicated One of the best tools to design acomplicated system is to present a model for that systeme steps associated with the design and implementationprocedure of the cipher algorithm regardless of the tripleclassification of the hash symmetric and asymmetricfunctions at the highest level are shown in Figure 1

Cryptography is one of the main information securitycomponents to transmit information from the sender to thereceiver using the most secure method [27] Design of thealgorithms has different requirements depending on itsapplication in the embedded or nonembedded system [28]To design a robust algorithm various technical and non-technical factors should be considered so that the designedalgorithm has sufficient robustness [29] On the contrary theeffective factors should be in a coherent model with logicalintegration so that their impact on each other can bemeasured and evaluated for instance in [30ndash33] variousalgorithms have been evaluated in terms of some parame-ters In fact designing a conceptual model for the cipheralgorithm requires considering all factors components and

indices that affect the design and implementation of thecipher algorithms Accordingly in the design of the cipheralgorithm there should be a balance between ldquoefficiencyrdquoand ldquoresourcesrdquo required for a specific security level [34]Considering the design and generation process of the cipheralgorithm and classification of factors components andindices the conceptual cybernetic supermodel is used fordesign and implementation Cybernetic is mainly focused onsystem performance and how they control their activitiesand communicate with their components erefore thecybernetic pattern might be a scientific basis for making thecipher algorithms targeted e cybernetic model of thecipher algorithms has four components of approachstrat-egy main process support process and control process einteractions of themain and support processes constitute thestructure of the cipher system ese interactions result in acomplicated diagram To overcome this complexity a lev-eled structure and mathematical facilities such as graph andmatrix are used Accordingly the general cybernetic modelfor the design and implementation of the cipher algorithmsis shown in Figure 2 is model is comprised of four sec-tions development approachstrategy process main processsupport process and control process e main process in-cludes cryptography algorithms e support process is di-vided into two general classes of hardware and software econtrol process includes controlling the design imple-mentation and controlling the outcome As mentioned thismodel is designed in the general level and its processes andcomponents are studied in detail in Section 3

Since the cipher algorithms have specific complexitiesthe component model is used to facilitate the processesis type of design prevents spending time on unnecessarydetails e strategic model for design of cipher algorithmsshould be presented at a level of detail that creates a trade-odd between ldquoinclusionrdquo and ldquoapplicabilityrdquo ldquoInclusionrdquoindicated including various cipher algorithms Accord-ingly considering the component extracted for the mainsupport and control section a cybernetic model can beused to design and implement cipher algorithms in threelevels as shown in Figure 3 According to this model thereis a bidirectional relationship between the main design andimplementation process and the support process at eachstep as a result of this relationship the support require-ments are demanded by the main process and provided bythe support process

31 Data Matrix of the Design Model Components Sincethere are a large number of extracted objectives or measuresin the design of cipher algorithms and some of them overlapor eliminating some of them causes no problem forachieving themain goals the criterion reductionmethod canbe used to eliminate some measures Accordingly in thisstudy the approach presented in [19] is used to reduce thenumber of measures In the feature reduction process ifeliminating one measure does not change the effective set ofthe problem it is unnecessary erefore after feature re-duction the component extraction process is carried out Inthis step three general classes of measures are extracted

Computational Intelligence and Neuroscience 3

including design and implementation objectives of cipher al-gorithms applications of cipher algorithms and imple-mentation methods of the cipher algorithms In the followingthe reduced measures are classified into the above classes and

modeling is carried out based on available measures of thesethree classes Accordingly based on the level-3 cyberneticmodel (Figure 3) the following three matrices are constitutedto design and implement the cipher algorithms

Table 1 e importance of information security needs in telemedicine technology in the field of information transmission

Data transfer Veryimportant Important No

idea Nonsignificant

Implement network protocols to ensure the transmission of information and checkits integrity 265 430 43 0

Establish a communication protocol to share information between local healthinstitutions 50 37 13 0

Encrypt important files and information 262 635 0 22Investigation of encryption mechanism by technical team of security assessor 556 328 152 0Use combinations of numbers and letters for encryption 268 25 68 0Use uppercase and lowercase letters for encryption to access remote networknetworks 50 24 174 87

Methods for controlling the integrity of application information 353 635 89 22

Hash Function

AsymmetricAlgorithms

SymmetricAlgorithms

ImplementationBasicApplication

Ciph

er T

ext

EvaluationPlai

n te

xt

Figure 1 Main process of design and implementation of cipher algorithms at the highest level

Cipher textRequest Plain text

Main Processes

Control Processes

Process

Product

Supportive Processes

Software

Hardware

CryptographyAlgorithms Policy

StrategyDevelopment

Figure 2 General schematic of the conceptual model of the cipher algorithm design

4 Computational Intelligence and Neuroscience

(i) e relationship matrix of support indices withmain design and implementation processes of thecipher algorithms since there are 13 support indicesin the level-3 model (Figure 3) and 4 steps in thelevel-1 model (Figure 2) a 13lowast4 matrix is consti-tuted to determine the relationship between themembers of these two processes where its rows arethe elements of the support process and its columnsare the quadruple elements of the design andimplementation of cryptography algorithms eelements of this matrix are between 0 and 10 asshown in Table 2 [35] e value of each elementrepresents the effectiveness of each support index oneach design and implementation step Analysis ofthis matrix and its modeling computation providesthe possibility for the policy-makers and developersto manage the resources required for each step ofdesign and implementation of the cryptographyalgorithms and use the available hardware andsoftware resources optimally

(ii) e relationship matrix of the objectives with themain design process of the cryptography algo-rithms considering the seven objectives extracted in

Section 2 and four steps in the main process todetermine the role of each step in achieving theseven objectives a 7lowast4 matrix is constituted Sincethe number of final resources used to extract data is814 papers technical reports and documents theelements of this matrix are between 0 and 814 asgiven in Table 2 In this matrix the value of eachelement represents the number of studies docu-ments and reports indicating the relationship be-tween two components An interesting point in thismatrix is that all algorithms implemented in thestudied documents are evaluated

(iii) e relationship matrix of the implementationtechniques and the design objectives of the cryp-tography algorithms considering the 29 extractedtechniques for cryptography algorithm imple-mentation and seven main objectives of the cryp-tography algorithms a 29lowast 7 matrix is constitutedto determine how much each technique is used forcryptography implementation to achieve each ob-jective (Table 3) e elements of this matrix arebetween 0 and 814 A part of this matrix is shown inTable 4

ICT

RampD

Financial

Rule amp Regulation

Standard

Human resourceEducation

OrganizationStructure

Management

Cultural

International relation

Infrastructure

Material

equipments

Hash Function

ControllerProcess

Product

Policy

Requ

est

Plai

n Te

xt

Cipher Text

Supportive

Main Processes

Software

Symmetric Algorithm

Asymmetric Algorithm

Hardware

Figure 3 e cybernetic model of design and implementation of cipher algorithms at level 3

Computational Intelligence and Neuroscience 5

311 Design of Cryptography Algorithms Using the Cyber-netic Supermodel Considering the above discussion andpresence of numerous indices and components in thedesign and implementation of the cryptography algorithmwhich make it a complex system identifying the rela-tionship between these indices and ranking them is anecessity Accordingly in this section the Q-analysis

method is used and the output of the three matrices isanalyzed According to the level three of the proposedcybernetic model the design and implementation supportprocess of the cryptography algorithms includes 13 com-ponents On the contrary the design and implementationsteps of the cryptography algorithms also include foursteps constituting a 13 lowast 4 matrix

Table 2 e 13-component interaction matrix of the support and the main design and implementation processes of cryptographyalgorithms

Main process Level 1Cryptography algorithms Level 2

Application eoretical basis Implementation Evaluation Level 3

Support process

Software

Culture 5 6 5 2 Organizationstructure 8 7 7 7

International and public relations 5 4 4 7 Financial resources 7 6 7 6

Human resources and education 8 10 10 9 Research and development 10 8 8 10

Rules 7 4 5 10 Management 9 7 8 7

FAVA 10 5 8 7 Standard 10 3 8 10

HardwareEquipment 8 6 7 8

Infrastructure 9 5 5 7Material 7 2 5 5

Table 3 e relationship matrix of the objectives with the design process of the cryptography algorithms

Main process objectives Application eoretical basis Implementation EvaluationSecurity 203 386 516 516Simplicity 5 23 38 38Resources 41 30 81 81Flexibility 15 28 45 45Scalability 6 11 22 22Speed 72 146 247 247Reliability 2 10 11 11

Table 4 A part of the interaction matrix between the cryptography algorithm implementation techniques and the seven objectives

Level 3 Security Simplicity Using resources Flexibility Scalability Speed Reliability

Support process

Avalanche effect 32 0 0 0 0 0 0Digital signature 1 0 0 0 0 0 0

Block size 1 1 0 0 2 0 0Image sharing 0 0 0 1 0 0 0

Parallel processing 0 0 0 1 12 0 0reshold technique 3 1 0 0 0 0 0

Data mining 1 0 0 0 0 0 0Binary tree 1 0 0 0 3 0 0

Cycle 5 2 0 0 12 0 0Multistage crypto 1 0 0 0 0 0 0Hybrid method 30 3 0 2 9 0 2

Hardware 0 0 7 0 0 0 0

6 Computational Intelligence and Neuroscience

32 Calculating the Incidence Matrix First the incidencematric is obtained based on the data matrix 4ndash1 is matrixrepresents the ldquoimpact of support indices on the main designprocess of the cryptography algorithmsrdquo e data matrix iscomprised of two sets D support indices set C and qua-druple design and implementation steps (Table 5) e in-cidence matrix calculated using the data matrix for α 70is represented in Table 6 By assigning different values to theparameter α difference incidence matrices are obtainederesults of the Q-analysis using C++ coding for α 70 aregiven in Figure 4

D d1 d2 d131113864 1113865

C c1 c2 c3 c41113864 1113865(1)

321 Geometric Representation Multidimensional proper-ties of the system are defined by a simple or complex setKD(C λ) such that the entities of the set D represent thesupport indices and entities of the set C represent thequadruple design and implementation steps of the cryp-tography algorithms

In the sample with α70 7 dis is as follows

d1

d2 c1 c2 c3 c41113864 1113865

d3 c41113864 1113865

d4 c1 c31113864 1113865

d5 c1 c2 c3 c41113864 1113865

d6 c1 c2 c3 c41113864 1113865

d7 c1 c41113864 1113865

d8 c1 c2 c3 c41113864 1113865

d9 c1 c3 c41113864 1113865

d10 c1 c3 c41113864 1113865

d11 c1 c3 c41113864 1113865

d12 c1 c41113864 1113865

(2)

e simplexes of σp(di) are also

σminus1 d1( 1113857 σ3 d2( 1113857 σ0 d3( 1113857 σ1 d4( 1113857 σ3 d5( 1113857 σ0 d3( 1113857 σ1 d7( 1113857

σ3 d8( 1113857 σ2 d9( 1113857 σ2 d10( 1113857 σ2 d11( 1113857 σ1 d12( 1113857 σ1 d13( 1113857 (3)

erefore the complex dimension is 3 In other wordsthe diagnosis classes d2 (structureorganization) d5 (humanresourceseducation) d6 (research and development) andd8 (management) have the largest dimension

33 CalculatingDimensions andQ-Link Q-link is defined asthe link between a subset with smallest interface between twosubsequent dis in the chain of d1 to dn Q-link between twosubsequent dis with α70 7 is

Table 5 Sets of dis and cis support indices based on the datamatrix

d1 Culture

d2 Structureorganizationd3 Internationalpublic relationsd4 Financiald5 Human resourceseducationd6 Research and developmentd7 Rulesd8 Managementd9 FAVAd10 Standardd11 Equipmentd12 Infrastructured13 MaterialC1 ApplicationC2 eoretical basisC3 ImplementationC4 Evaluation

Table 6 e incidence matrix of the support indicesrsquo impact of thedesign steps of the cryptography algorithms with α 70

C1 C2 C3 C4

d1 0 0 0 0d2 1 1 1 1d3 0 0 0 1d4 1 0 1 0d5 1 1 1 1d6 1 1 1 1d7 1 0 0 1d8 1 1 1 1d9 1 0 1 1d10 1 0 1 1d11 1 0 1 1d12 1 0 0 1d13 1 0 0 0

Computational Intelligence and Neuroscience 7

σminus1 d1( 1113857 σ3 d2( 1113857⟶ 1 σ3 d2( 1113857 σ0 d3( 1113857⟶ 0 σ0 d3( 1113857 σ1 d4( 1113857⟶ minus 1

σ1 d4( 1113857 σ3 d5( 1113857⟶ 1 σ3 d5( 1113857 σ3 d6( 1113857⟶ 3 σ3 d6( 1113857 σ1 d7( 1113857⟶ 1

σ1 d7( 1113857 σ3 d8( 1113857⟶ 1 σ3 d8( 1113857 σ2 d9( 1113857⟶ 2 σ2 d9( 1113857 σ2 d10( 1113857⟶ 2

σ2 d10( 1113857 σ2 d11( 1113857⟶ 2 σ3 d11( 1113857 σ1 d12( 1113857⟶ 1 σ1 d12( 1113857 σ0 d13( 1113857⟶ 0

(4)

e maximum link dimension is 3 indicating the re-lationship between the diagnosis classes

331 Calculating the Structure Vectors As mentioned inthe definitions the vector Qq is a simplification basiscreated to eliminate the additional impacts in the equiv-alent simplex sets e maximum complex dimension withα70 7 is 3 erefore the first structure vector based onthe output is

Dimension 3 2 1 0 (5)

e second structure vector P is

dimensions 3 2 1 0

P Pdim3 Pdim2 Pdim1 Pdim0( 1113857

P 4 7 10 12( 1113857

(6)

where Pq is the number simplexes greater than or equal to qin the set K in which P is the number of simplex linkrepetitions (support indices) in the quadruple design andimplementation steps of the cryptography algorithms Basedon the values of these two vectors it is seen that the rela-tionship between the support indices and the quadrupledesign and implementation steps of the cryptography al-gorithms is high is issue indicates the role of supportcomponents in the design and implementation of the

cryptography algorithms which should be consideredseriously

332 Calculating the Obstruction or Flexibility VectorQlowastK represents the number of structural obstructions forsimplex interactions in dimension k

Qlowast

Q minus I⟶ Qlowast

1 1 1 11113858 1113859 minus 1 1 1 11113858 1113859⟶ Qlowast

0 0 0 01113858 1113859

(7)

As can be seen there is no obstruction in any of thecommunication levels indicating that there is a significantrelationship between the support components at eachequivalence class

333 Calculating Irregularity e value of (eccprime(σ)) iscalculated using the Chinese method e results for α70

7 are shown in Table 7 e calculated irregularity valueshows that the indices d2 (structureorganization) d5 (hu-man resourceseducation) d6 (research and development)and d8 (management) affect other indices

334 Calculating Complexity Also the results ofQ-analysiscan be used to describe structure complexity According toequations (4)ndash(9) and for α = 7 the complexity measure is

-1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1

3 0 1 3 3 1 3 2 2 2 1

0 -1 0 0 0 0 0 0 0 0

1 1 1 0 1 1 1 1 0

3 3 1 3 2 2 2 1

3 1 3 2 2 2 1

3 2 2 2 1

2 2 2 1

2 2 1

2 1

1

1 1 1 1 1 1

-1

-1

0

0

0

0

0

0

0

0

0

0

0q Q3210

11

11

SETSX2 X5 X6 X8X2 X5 X6 X8 X9 X10 X11X2 X4 X5 X6 X7 X8 X9 X10 X11 X12X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13

Figure 4 Implementation results of the model for support components of the cryptography algorithmsrsquo design with α 70

8 Computational Intelligence and Neuroscience

Q 1 1 1 1( 1113857

ψ(K) 2(1 + 2 + 3 + 4)

(4lowast 5)1113890 1113891 1

(8)

Since in the above model there is no obstruction amongthe component of the equivalent class it was expected thatthe complexity of the support components is not high andthe obtained complexity index of 1 verifies this expectatione system complexity for different alpha-cuts is shown inFigure 5

335 Ranking the Support Components of the Design andImplementation of the Cryptography Algorithms e resultsof using A-analysis are shown in Table 7 e connectionstrength of the factors in one group is specified with alpha-cut erefore the support components are grouped in 5levels Each level describes the priority and importance of thegroup in developing the cryptography algorithms In Fig-ure 6 the ranking pyramid of the support components usingQ-analysis is shown To allocate proper resources thecomponents existing in higher levels of the pyramid (Fig-ure 6) are of higher priority

336 Validation of the Results In this section the resultsof the cybernetic model and Q-analysis for supportcomponentsrsquo ranking are compared with the results re-ported in the global cybersecurity index (GCI) in 20152017 and 2018 presented by ITU [36ndash39] e GCI reportsare focused on five indices including ldquolegal cases orga-nization necessities technical issues capacity buildingand cooperationrdquo and the subindices include legaltechnical organization capacity building and coopera-tion According to the presented indices and subindices itis clear that the ldquorulesrdquo ldquostandardrdquo ldquoresearch and de-velopmentrdquo ldquoeducationrdquo and ldquomanagementrdquo are ofhigher priority in security establishment Although ourresearch is more skilled and detailed compared to the GCIreports but the results verify our findings Also Table 8shows the ranking of the support components obtainedusing Q-analysis

337 Executing the Model and Analyzing the Results of theObjectivesrsquo Impact on Design Steps of the Cryptography Al-gorithmsrsquo Matrix In this section the role of the sevencomponents on the quadruple design and implementationsteps of the cryptography algorithm is analyzed with Q-analysis Using the Q-analysis method and the 7lowast4 matrixobtained from the relationship of the cryptography algo-rithmsrsquo design objectives on their quadruple steps theirindices are ranked

338 Calculating the Incidence Matrix and Executing theModel First the incidence matrix is obtained for the datamatrix shown in Figure 2 It can be seen in Table 9 that eachelement of the two sets represents which indice e inci-dence matrix calculated from the data matrix for α5 40 is

given in Table 10 e results of implementing the model forα5 40 are given in Figure 7

4 Geometric Representation

In the sample with α5 40 dis are

d1 c1 c2 c3 c41113864 1113865

d2

d3 c1 c3 c41113864 1113865

d4 c3 c41113864 1113865

d5

d6 c1 c2 c3 c41113864 1113865

d7

(9)

e simplexes of σp(di) are also

σ3 d1( 1113857σminus1 d2( 1113857σ2 d3( 1113857σ1 d4( 1113857σminus1 d5( 1113857σ3 d6( 1113857σminus1 d7( 1113857

(10)

erefore the complex dimension is 3 In other wordsthe discriminant classes d1 (security) and d6 (speed) have thelargest dimension

41 Calculating the Structure Vectors As mentioned thevector Qq is a simplification basis to eliminate the additionaleffects in the set of equivalent simplexes e maximumcomplex dimension for α5 40 is 3 erefore the firststructure vector based on the output is

Dimension 3 2 1 1

Q 1 1 1 1( 1113857

Dimension 3 2 1 0

P 2 3 4 7( 1113857

(11)

e second structure vector P is

P 2 3 4 7( 1113857 (12)

42 Calculating the Obstruction Vector or InflexibilityQlowastK represents the number of structural obstructions forsimplex interactions in dimension k which is as follows forα5 40

1 1 1 1 1332

005

115

225

com

plex

ity

alpha cutα=5 α=6 α=7 α=8 α=9 α=10

Figure 5 System complexity of the support indices of design andimplementation of the cryptography algorithms for different alpha-cuts

Computational Intelligence and Neuroscience 9

Qlowast

Q minus I⟶ Qlowast

2 1 1 31113858 1113859 minus 1 1 1 11113858 1113859⟶ Qlowast

0 0 0 01113858 1113859

(13)

According to the obtained values it can be concludedthat there are no structural obstructions among the main

design indices of the cryptography algorithms at ismultiple objectives are considered simultaneously by thecryptography algorithm designers

Table 7 e irregularity of the data matrix parameters for α= 7

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d2 d5 d6 d8 qi 3 2 1 364 3 061d9 d10d11 qi 2 1 064 3 011d4 d7 d12 qi 1 014 3 002

human resourceRampD

rules FAVA standard

management infrastructure

organizationstructure equipmentinternational and public relations finanacial

resources and material

Figure 6 e ranking pyramid of the support components using Q-analysis

Table 8 Ranking of the support components using Q-analysis

Relationship of the support components with the design and implementation of the cryptography algorithms (q 0)No relationship α 0 complete relationship α 100

Human resource RampD rules FAVA and standard α100 10Management and infrastructure α90 9Organizationstructure and equipment α80 8International and public relations financial resources and material α70 7Culture α60 6All components α60 5

Table 9 Set of dis and cis the objectivesrsquo indices

eoretical basis C2 Reliability d7 Scalability d5 Resources d3 Security d1Implementation C3 Application C1 Speed d6 Flexibility d4 Simplicity d2Evaluation C4

Table 10 e incidence matrix of the design objectives of thecryptography algorithms with α= 40

c1 c2 c3 c4

d1 1 1 1 1d2 0 0 0 0d3 1 0 1 1d4 0 0 1 1d5 0 0 0 0d6 1 1 1 1d7 0 0 0 0

-1 -1 -1 -1 -1

2 1 -1 2

1 -1 1

-1 -1

3

3 -1 2 1 -1 3 -1

-1

-1

-1

-1

-1

-1q Q3210

11

11

SETSX1X6X1X3X6X1X3X4X6X1X3X4X6

Figure 7 e results of executing the model for design objectivesrsquoindices with α 40

10 Computational Intelligence and Neuroscience

43 Calculating Irregularity e results of applying theChinese method (ecclsquo(σ)) for calculating irregularity forα5 40 are shown in Table 11 e calculated irregu-larity value shows that indices d1 (security) and d6(speed) have received more attention compared to otherindices

44 Calculating Complexity Structural complexity for asample with α5 40 is

dimensions 3 2 1 0

Q Qdim3 Qdim2 Qdim1 Qdim0( 1113857

Q 1 1 1 1( 1113857

ψ(K) 2(1 + 2 + 3 + 4)

(4lowast 5)1113890 1113891 1

(14)

Since in the above model there was no obstructionamong the components of the equivalent classes in any of thecommunication levels it was expected that there is not ahigh complexity among objectives of the cryptography al-gorithm design and the complexity index of 1 verified thisexpectation e system complexity for different alpha-cutsis shown in Figure 8 As can be seen the system complexity is1 for all significant values of alpha

45 Prioritizing the Parameters of the Main Objectives of theCryptography Algorithm Design Using Q-AnalysisAccording to the analysis results the parameters can beclassified into 5 levelsese levels are shown in Table 12 andFigure 9 Each level indicates a priority and importance ofthe group in the development of cryptography algorithmsTo allocate proper resources the components at the higherlevels of the pyramid are of higher priority As can be seenthree objectives of security speed and optimal usage ofresources have the highest priority for the design of cryp-tography algorithms

46 Executing the Model and Analyzing the Results for theCryptography Algorithm Implementation Techniques einteraction matrix between the 29 extracted techniques andthe sevenmain objectives is shown in Figure 10e purposeof this section is to rank the cryptography algorithmimplementation techniques to achieve the goals of interest

47Calculating the IncidenceMatrixandExecuting theModele data matrix A is comprised of two sets e set Drepresents the employed techniques and the set C representsthe seven main design objectives (Table 13)

D d1 d2 d291113864 1113865

C c1 c2 c71113864 1113865(15)

Some parts of the incidence matrix calculated from thedata matrix A for α3 are shown in Table 14 and the resultsof theQ-analysis model for α3 24 are shown in Figure 11

48 Geometric Representation In the sample with α3 24the dis are

d1 c11113864 1113865

d2

d3

d4

d5

d6

d7

d8

d9

d10

d11 c11113864 1113865

d12

d13 c51113864 1113865

d14

d15 c1 c4 c51113864 1113865

d16

d17

d18

d19 c11113864 1113865

d20

d21

d22 c31113864 1113865

d23

d24

d25

d26

d27 c11113864 1113865

d28

d29

(16)

e simplexes of σp(di) are

Computational Intelligence and Neuroscience 11

σ0 d1( 1113857

σminus1 d2( 1113857

σminus1 d3( 1113857

σminus1 d4( 1113857

σminus1 d5( 1113857

σminus1 d6( 1113857

σminus1 d7( 1113857

σminus1 d8( 1113857

σminus1 d9( 1113857

σminus1 d10( 1113857

σ0 d11( 1113857

σminus1 d12( 1113857

σ0 d13( 1113857

σminus1 d14( 1113857

σ2 d15( 1113857

σminus1 d16( 1113857

σminus1 d17( 1113857

σminus1 d18( 1113857

σ0 d19( 1113857

σminus1 d20( 1113857

σminus1 d21( 1113857

σ0 d22( 1113857

σminus1 d23( 1113857

σminus1 d24( 1113857

σminus1 d25( 1113857

σminus1 d26( 1113857

σ0 d27( 1113857

σminus1 d28( 1113857

σminus1 d29( 1113857

(17)

erefore the complex dimension is 2 In other wordsthe discriminant class d15 (basic sciences) has the largestdimension

49 Calculating the Dimensions and the Q-Link In the alphadefined for cryptography algorithm implementation tech-niques to achieve the defined goals the obtained q-linkshows that these techniques are relatively independent al-though there is a weak relationship between some tech-niques Q-link in the samples with α3 24 between eachtwo subsequent di is as follows

σ0 d1( 1113857 σminus1 d2( 1113857⟶ minus 1

σminus1 d2( 1113857 σminus1 d3( 1113857⟶ minus 1

σ0 d3( 1113857 σminus1 d4( 1113857⟶ minus 1

σminus1 d4( 1113857 σminus1 d5( 1113857⟶ minus 1

σminus1 d5( 1113857 σminus1 d6( 1113857⟶ minus 1

σminus1 d6( 1113857 σminus1 d7( 1113857⟶ minus 1

σminus1 d7( 1113857 σminus1 d8( 1113857⟶ minus 1

σminus1 d8( 1113857 σminus1 d9( 1113857⟶ minus 1

σminus1 d9( 1113857 σminus1 d10( 1113857⟶ minus 1

σminus1 d10( 1113857 σ0 d11( 1113857⟶ minus 1

σ0 d11( 1113857 σminus1 d12( 1113857⟶ minus 1

σminus1 d12( 1113857 σ0 d13( 1113857⟶ minus 1

σ0 d13( 1113857 σminus1 d14( 1113857⟶ minus 1

σminus1 d14( 1113857 σ2 d15( 1113857⟶ minus 1

σ2 d15( 1113857 σminus1 d16( 1113857⟶ minus 1

σminus1 d16( 1113857 σminus1 d17( 1113857⟶ minus 1

σminus1 d17( 1113857 σminus1 d18( 1113857⟶ minus 1

σminus1 d18( 1113857 σ0 d19( 1113857⟶ minus 1

σ0 d19( 1113857 σminus1 d20( 1113857⟶ minus 1

σminus1 d20( 1113857 σminus1 d21( 1113857⟶ minus 1

σminus1 d21( 1113857 σ0 d22( 1113857⟶ minus 1

σ0 d22( 1113857 σminus1 d23( 1113857⟶ minus 1

σminus1 d23( 1113857 σminus1 d24( 1113857⟶ minus 1

σminus1 d24( 1113857 σminus1 d25( 1113857⟶ minus 1

σminus1 d25( 1113857 σminus1 d26( 1113857⟶ minus 1

σminus1 d26( 1113857 σ0 d27( 1113857⟶ minus 1

σ0 d27( 1113857 σminus1 d28( 1113857⟶ minus 1

σminus1 d28( 1113857 σminus1 d29( 1113857⟶ minus 1

(18)

Table 11 Irregularity of the cryptography objectivesrsquo parameters in the data matrix A for α5

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d1 and d6 qi 3 2 and 1 433 3 072d3 qi 2 and 1 133 3 026d4 qi 1 033 3 006

11117

1 1 1

09095

1105

11115

12

com

plex

ity

alpha cutα=2 α=3 α=5 α=15 α=25

Figure 8 System complexity of the cryptography algorithm designobjectives for different alpha-cuts

12 Computational Intelligence and Neuroscience

410 Calculating the Structure Vectors erefore the firststructure vector based on the software output is

Q 1 1 2( 1113857 (19)

e second structure vector P is

P 1 1 29( 1113857 (20)

Since the large values of P of higher dimensions dem-onstrate more links the second structure vector calculatedfor the alpha of interest shows that the relationship betweenthe cryptography algorithm implementation techniques isminimum

411 Calculating the Obstruction or Inflexibility Vectore obstruction vector (Qlowast) for a sample with α3 24 is

Table 12 Ranking the objectivesrsquo parameters considering different cuts in Q-analysis

e equivalence class for the parameters with minimum relation level (q 0) αSecurity α50 244Speed α30 244Resources α15 122Flexibility α5 40Simplicity-scalability α2 16(All parameters)

securityspeed

optimal usage of resources

flexibility

simplicity-scalability

Figure 9 Prioritizing the design and implementation objectives of the cryptography algorithms using Q-analysis

securityspeed

optimal usage of resources

flexibility

simplicity-scalability-reliability

Figure 10 Prioritizing the design and implementation objectives of the cryptography algorithms using Q-analysis

Table 13 Set of dis and cis the techniques employed for implementing cryptography algorithms

Flexibility C4 Fuzzy logic d25 Storage space d17 Cycle d9 Avalanche effect d1Scalability C5 Software d26 Clustering d18 Multiple step d10 Digital signature d2Speed C6 Steganography d27 Key d19 Hybrid method d11 Block size d3Reliability C7 Music harmony d28 Graph d20 Hardware d12 Image sharing d4

Artificial intelligence d29 Characteristic oriented d21 Hardware d13 Parallel processing d5Security C1 Energy consumption d22 Occupation area d14 reshold technique d6Simplicity C2 Bandwidth consumption d23 Basic science d15 Data mining d7Resources C3 Memory consumption d24 Compression d16 Binary tree d8

Table 14 e incidence matrix of the cryptography algorithmimplementation techniques with α 24

C1 C2 C3 C4 C5 C6 C7

d1 1 0 0 0 0 0 0d2 0 0 0 0 0 0 0d3 0 0 0 0 0 0 0d4 0 0 0 0 0 0 0d5 0 0 0 0 0 0 0d6 0 0 0 0 0 0 0d7 0 0 0 0 0 0 0d8 0 0 0 0 0 0 0d9 0 0 0 0 0 0 0d10 0 0 0 0 0 0 0d11 1 0 0 0 0 0 0d12 0 0 0 0 0 0 0

Computational Intelligence and Neuroscience 13

Qlowast

Q minus I⟶ Qlowast

1 1 21113858 1113859 minus 1 1 11113858 1113859⟶ Qlowast

0 0 11113858 1113859

(21)

e value of QlowastK represents the number of structurallimitations or obstructions for the cryptography techniquesrsquointeraction at dimension k As can be seen from the cal-culated obstruction vector there is a significant relationshipat some levels

412 Calculating Irregularity Irregularity is the integrationdegree of a cryptography method in the total complex Mea-suring irregularity (eccrsquo) for α3 24 is shown in Table 15 Ascan be seen irregularity for the discriminant class d15 indicatesthat this technique is isolated from other techniques

413 Calculating Complexity Structure complexity for asample with α3 24 is

Q 1 1 2( 1113857

ψ(K) 2(2 + 2 + 3)

(3lowast 4)1113890 1113891 117

(22)

e system complexity for different values of alpha isshown in Figure 12 As can be seen the total complexity ofthe system for large values alpha tends to stability

414 Ranking the Cryptography Algorithm ImplementationTechniques ese techniques can be classified into 5 levelsusing the analysis of the obtained results ese levels can beseen in Table 16 and Figure 13 e link power of the factorsin one group is specified with alpha Each level describespriority and importance of the group in the development ofcryptography algorithms Accordingly six methods of basicscience key management using hardware methods usingsteganography Avalanche effect and using hybrid method

q Q21o

112

SETSX15X15X1X11X15X19X27X13 amp X22

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-10-1-1-10-1-1-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-10-1-1-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

2-10-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1

-1-1-1-1-1-1

-1-1-1-1-1

-1-1-1-1

-1-1-1

-1-10

Figure 11 Results of executing the model for cryptography algorithm implementation techniques with α3 24

14 Computational Intelligence and Neuroscience

are the most important cryptography algorithm imple-mentation techniques to achieve the main objectives

5 Conclusion

In this study the design and implementation model of thecryptography algorithms is designed in three levels and itseffective components are extracted To organize the com-ponents in addition to elimination of the additional com-ponents through the measure reduction algorithm a properclassification is applied to examine the mutual effects Afterclassification three 13lowast 4 7lowast 4 and 29lowast 7 matrices areconstituted and the model is implemented on these threematrices To implement the designed model Q-analysis isused Accordingly for support indices five indices of highpriority include human resources research and develop-ment management organizationstructure and equipmentFor the algorithm design objectives index five high priorityindices include security speed optimal usage of resources

and simplicity For the indices related to implementationtechniques of the cryptography algorithms the most appliedtechniques for achieving the determined objectives includeusing basic science hardware and software methods usingkeymanagement hybrid method steganography Avalancheeffect and using artificial intelligence In the future work wewill consider the following cases

A plan should be formulated according to the prioritiesAccording to the outputs of this study and the presented pri-orities the following topics can be investigated in future studies

(1) Presenting a comprehensive model for generatingvarious cryptography algorithms based on the pri-orities of interest

(2) Examining and formulating a model about the roleof basic science for design and implementation ofcryptography algorithms considering the significantrole of ldquobasic sciencerdquo in implementation of thecryptography algorithms and about the role ofschools and universities

(3) Presenting a model for design and implementationof cryptography algorithms in the IoT infrastructurewith optimal resource usage

Data Availability

e data used to support the findings of the study areavailable from the corresponding author upon request

Conflicts of Interest

e authors declare that there are no conflicts of interestregarding the publication of this article

Table 15 Irregularity of the cryptography algorithm implementation techniques for α3 24

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d15 qi 2 1 3 2 1- qi 1 1 2 033

11117

1 1 1

091

1112

com

plex

ity

alpha-cutα=2 α=3 α=5 α=15 α=25

Figure 12 System complexity of cryptography algorithm implementation techniques for different alpha-cuts

Table 16 Ranking the implementation techniques of the cryptography algorithms considering various alpha-cuts using Q-analysis

Equivalence classes for the parameters with minimum relationship level (q 0) αBasic sciences α20 163Key α10 81Hardware and steganography α5 40Avalanche effect and hybrid method α3 24Artificial intelligence software occupation area energy consumption memory consumption α2 16

basic sciences

key

Avalanche effect hybrid method

hardware steganography

artificial intelligence soware occupationarea energy consumption memory consumption

Figure 13 Prioritizing the implementation techniques of thecryptography algorithms using Q-analysis

Computational Intelligence and Neuroscience 15

Acknowledgments

is work was partially supported by Iran National ScienceFoundation (INSF) under contract No 9653979

References

[1] D Nilesh and M Nagle ldquoe new cryptography algorithmwith high throughputrdquo in Proceedings of the 2014 Interna-tional Conference on Computer Communication and Infor-matics pp 3ndash7 Coimbatore India January 2014

[2] S Ravi P Kocher R Lee G McGraw and A RaghunathanldquoSecurity as a new dimension in embedded system designrdquo inProceedings of the 41st annual conference on Design auto-mation-DAC lsquo04 pp 753ndash760 San Diego CA USA March2004

[3] P Kuppuswamy and S Q Y A Khalidi ldquoHybrid encryptiondecryption technique using new public key and symmetric keyalgorithmrdquo International Journal of Information and Com-puter Security vol 6 no 4 pp 372ndash382 2014

[4] S Bhat and V Kapoor ldquoSecure and efficient data privacyauthentication and integrity schemes using hybrid cryptog-raphyrdquo International Conference on Advanced ComputingNetworking and Informatics vol 870 pp 279ndash285 2019

[5] D Ott and C Peikert ldquoAnd other workshop participantsldquoidentifying research challenges in post quantum cryptog-raphy migration and cryptographic agilityrdquo 2019 httparxivorgabs190907353

[6] P Manickam K Shankar E Perumal M Ilayaraja andK Sathesh Kumar ldquoSecure data transmission through reliablevehicles in VANET using optimal lightweight cryptographyrdquoin Advanced Sciences and Technologies for Security Applica-tions pp 193ndash204 Springer New York NY USA 2019

[7] S Prakash and A Rajput ldquoHybrid cryptography for securedata communication in wireless sensor networksrdquo Advancesin Intelligent Systems and Computing vol 696 pp 589ndash5992018

[8] A A Soofi I Riaz and U Rasheed ldquoAn enhanced vigenerecipher for data securityrdquo International Journal of Scientific ampTechnology Research vol 4 no 8 pp 141ndash145 2015

[9] N Shamsuddin M Syafiqah and S Ali Pitchay ldquoImple-menting location-based cryptography on mobile applicationdesign to secure data in cloud storagerdquo Journal of PhysicsConference Series IOP Publishing vol 1551 no 1 2020

[10] L Jiao H Yonglin and D Feng ldquoStream cipher designs areviewrdquo Science China Information Sciences vol 63 no 3pp 1ndash25 2020

[11] NIST ldquoNIST cryptographic standards and guidelines devel-opment processrdquo Inside NIST vol 27 2016

[12] G Alagic J Alperin-Sheriff D Apon et al Status Report onthe First Round of the NIST Post-Quantum CryptographyStandardization Process National Institute of Standards andTechnology Gaithersburg MD USA pp 1ndash27 2019

[13] I Damaj and S Kasbah ldquoAn analysis framework for hardwareand software implementations with applications from cryp-tographyrdquo Computers amp Electrical Engineering vol 69pp 572ndash584 2018

[14] M S Croock Z A Hassan and S D Khudhur ldquoAdaptive keygeneration algorithm based on software engineering meth-odologyrdquo International Journal of Electrical and ComputerEngineering vol 11 no 1 p 589 2021

[15] K Sha T A Yang W Wei and S Davari ldquoA survey of edgecomputing-based designs for iot securityrdquo Digital Commu-nications and Networks vol 6 no 2 pp 195ndash202 2020

[16] M Yahuza M Y I B Idris A W B A Wahab et alldquoSystematic review on security and privacy requirements inedge computing state of the art and future research oppor-tunitiesrdquo IEEE Access vol 8 pp 76541ndash76567 2020

[17] A Bhowmik J Dey and S Karforma A Novel Double TierCryptographic System (nDTCS) to Reinforce Patientsrsquo Privacyin Contemporary COVID-19 Telemedicine Springer NewYork NY USA 2021

[18] N Gupta User Integrity Protection Security Model forEnhanced Telemedicine for Healthcare Networks 2020httpsosfio69jdq

[19] J Dey A Bhowmik A Sarkar S Karforma andB Chowdhury Cryptographic Engineering on COVID-19Telemedicine An Intelligent Transmission through RecurrentRelation Based Session Key Springer New York NY USA2021

[20] V Hosseinian H Ayatollahi H Haghani and E MehraeenldquoRequirements of information security in a telemedicinenetwork review of IT managersrsquo opinionrdquo Journal of Para-medical Sciences amp Rehabilitation vol 4 no 2 pp 31ndash402015

[21] D Karaoglan Altop A Levi and V Tuzcu ldquoDeriving cryp-tographic keys from physiological signalsrdquo Pervasive andMobile Computing vol 39 pp 65ndash79 2017

[22] O Uzunkol and M S Kiraz ldquoStill wrong use of pairings incryptographyrdquo Applied Mathematics and Computationvol 333 pp 467ndash479 2018

[23] S Picek D Jakobovic J F Miller L Batina and M CupicldquoCryptographic Boolean functions one output many designcriteriardquo Applied Soft Computing vol 40 pp 635ndash653 2016

[24] R Saha and G Geetha ldquoSymmetric random function gen-erator (SRFG) a novel cryptographic primitive for designingfast and robust algorithmsrdquo Chaos Solitons amp Fractalsvol 104 pp 371ndash377 2017

[25] A Achuthshankar and A Achuthshankar ldquoA novel sym-metric cryptography algorithm for fast and secure encryp-tionrdquo in Proceedings of the 2015 IEEE 9th InternationalConference on Intelligent Systems and Control (ISCO) pp 4ndash9Coimbatore India January 2015

[26] D Ott and C Peikert ldquoIdentifying research challenges in postquantum cryptography migration and cryptographic agilityrdquo2019 httparxivorgabs190907353

[27] W Liu B Ying H Yang and H Wang ldquoAccurate modelingfor predicting cryptography overheads on wireless sensornodesrdquovol 2 pp 997ndash1001 in Proceedings of the 2019 11thInternational Conference on Advanced CommunicationTechnology vol 2 pp 997ndash1001 IEEE Bangalore IndiaFebruary 2019

[28] S Feizi A Ahmadi and A Nemati ldquoA hardware imple-mentation of simon cryptography algorithmrdquo in Proceedingsof the 2014 4th International Conference on Computer andKnowledge Engineering (ICCKE) pp 245ndash250 IEEEMashhad Iran 2014

[29] E ambiraja G Ramesh and R Umarani ldquoA survey onvarious most common encryption techniquesrdquo 2012 httpwwwijarcssecom

[30] A Gupta and N K Walia ldquoCryptography Algorithms areviewrdquo International Journal of Engineering Research andDevelopment vol 2 no 2 pp 1667ndash1672 2014

[31] R Bhanot and R Hans ldquoA review and comparative analysis ofvarious encryption algorithmsrdquo International Journal of Se-curity and Its Applications vol 9 no 4 pp 289ndash306 2015

[32] S Chandra S Paira S S Alam and G Sanyal ldquoA com-parative survey of symmetric and asymmetric key

16 Computational Intelligence and Neuroscience

cryptographyrdquo in Proceedings of the 2014 InternationalConference on Electronics Communication and Computa-tional Engineering (ICECCE) pp 83ndash93 Hosur India No-vember 2014

[33] F Maqsood M Ahmed M Mumtaz and M Ali ldquoCryp-tography a comparative analysis for modern techniquesrdquoInternational Journal of Advanced Computer Science andApplications vol 8 no 6 p 6 2017

[34] S B Sadkhan and A O Salman ldquoA survey on lightweight-cryptography status and future challengesrdquo in Proceedings ofthe International Conference on Advances in SustainableEngineering and Applications pp 105ndash108 Wasit Iraq June2018

[35] N V oai ldquoCriteria and dimension reduction of linearmultiple criteria optimization problemsrdquo Journal of GlobalOptimization vol 52 no 3 pp 499ndash508 2012

[36] C E Shannon ldquoCommunication theory of secrecy systems1945rdquo MD Computing vol 15 no 1 pp 57ndash64 1998

[37] L J Fennelly M Beaudry and M A Perry Security in 2025[38] ITU Global Cybersecurity Index 2018 2019[39] ITU Global Cybersecurity Index (GCI) 2017 2017

Computational Intelligence and Neuroscience 17

algorithm should be systematic comprehensive and stagedAll required components of the information security shouldbe considered in an excellence pattern in terms of technicalorganization procedural and humanitarian aspects Iden-tifying new cryptographic challenges such as post-quantumcryptography and its agility mobile applications robustnessof algorithms and the role of implementation methods toachieve the above goals can be implemented in a compre-hensive model [5ndash7] Providing all these requirements si-multaneously in the design of an algorithm is difficult andsometimes impossible If contradictory objectives are con-sidered for formulation of an objectiveobjectives of analgorithm most algorithms might be broken and if attackerhas sufficient time motivation and resources he can trackthe information [8] Accordingly presenting a model tomake the cryptography algorithm design targeted is veryimportant erefore in this study an approach is presentedto design a leveled model of cryptography algorithms usingthe cybernetic method e rest of this paper is structured asfollows In the second section the literatures review ispresented In the third section the cybernetic methodologyis described and a model is specified for design andimplementation of cryptography algorithms based onextracted indices In the fourth section the proposed modelis used to examine the design of the cryptography algorithmusing the cybernetic supermodel Finally the proposedapproach is evaluated using the Q-analysis method

2 Literature Review

In [9] a security approach based on cryptography has beenpresented through examining the security issues in mobiledevices and the available solutions Also it is mentioned thatasymmetric cryptography is not a proper option for securingthe resource-limited infrastructures such as IoT due to highcomplexity of the design and implementation On thecontrary employing symmetric algorithms has other se-curity issues Accordingly it has studied the design ofcryptography algorithms based on position To this end anapproach based on the AES algorithm and position of anefficient cryptography approach has been designed In thisapproach the application diagram is described and the useroperation is studied In the following the operation flows ofthe system are described Finally it was evaluated that se-curity can be increased through employing this approach In[10] the design of stream cipher algorithms has beenstudied It has been mentioned that stream cipher is one ofthe essential branches of symmetric cryptography whichrequires limited hardware resources for execution ere-fore considering the development of the communicationtechnologies the need to these algorithms is increasingAccordingly in the following the design procedures andperformance of various encryption algorithms includingNFSR eStram FCSR and Panama are presented throughdescribing the main requirements of the cryptography al-gorithm design e main purpose of this study is to presenta perspective of stream cipher algorithm design and theirperformance In 2016 NIST published a document calledcipher standard and development instructions [11]

Transparency openness balance accuracy technical meritglobal acceptability usability continuous improvement andinnovation and intellectual property (IIP) are the guidanceprinciples of NIST cipher standards and developmentprocedures Also NIST has started a procedure to requestevaluate and standard of one or multiple public key cipheralgorithms robust against quantum attacks [12] In [11]lifecycle management processes and policies of cipherstandard have been presented where its main principlesinclude identifying and evaluating the needs announcingthe userrsquos intention on a standard or instruction consideringthe requirements and solutions defining a specific programand procedure and design and development of a standardand evaluating and maintaining the standard In [13] ananalytical framework has been presented to hardware andsoftware implementation using cipher programs that verifiesan integrated statistical framework which can implement theclassified algorithms successfully based on a combination ofheterogeneous hardware features and their software appli-cations e model presented in this paper includes six el-ements of goal input activities output outcomes andperformance In [14] software engineering methodologieshave been used to propose an adaptive approach for pre-senting a robust cipher key generation algorithm etechnique used in this method is based on self-checkingprocedures that can detect the system-level errors ere-fore it can be used to check the security keys generated viaemploying random factors ese factors have been pre-sented in the NIST evaluation results In this softwaremethod the values of the random factors are smaller thanthe acceptance values and the key is generated when a validvalue is detected e generated keys are generated throughshift register and SIGBA technique e evaluation resultsindicate the efficiency of the presented approach in gener-ating valid cipher keys In [15 16] security issues of mobiledevices and processing infrastructure including mobilecomputing and edge computing have been studied It hasbeen concluded the importance of cipher algorithms andnecessity of employing new models consider the complexityof these infrastructures Bhowmik et al [17] focused onsecurity issues in telemedicine and introduced a double-tier(nDTCS) encryption system Accordingly this solution hasproposed a modified logistic map and a congruence-basedsecurity model to secure telemedicine medical transactionsTwo keys have been used for the encryption and decryptionprocess intermediate key and session key e results of theevaluation indicate the effectiveness of the solution in orderto secure the information through the proposing method[18] In order to protect telemedicine communications a keyexchange solution is proposed by improving the Dif-fiendashHellman cryptographic algorithm In this method arandomized key generation is used to generate the key eproposed solution is naturally safe and reliable due to the useof the DiffiendashHellman algorithm so there is no need forrecalculations or key reversal e results of the evaluationalso indicate that the proposed method is safe againstguessing key attacks In [19] an intelligent and securedtransmission security solution for heart disease reportsbased on session key-based methods is presented For this

2 Computational Intelligence and Neuroscience

purpose matrix confusion operations are used e inno-vation of this solution is in the process of matrix transferwhich is transmitted in the form of a number of cardiologistsin particular Finally the efficiency of the proposed methodis evaluated with regard to cryptographic engineeringtransparency and strength e results indicate that thismethod provides more security in the medical data trans-mission process Hosseinian et al [20] examined the im-portance of information security needs in telemedicinetechnology in the field of information transmission In thisstudy the data collection tool was a questionnaire that wasdesigned based on the criteria of the Association of Infor-mation Management and Health Care Systems (HMISS) inthe field of telemedicine network security and securitystandards of the American Telemedicine Association isquestionnaire has been calculated separately based on ascore of 1 to 5 Table 1 summarizes the importance of eachsection

3 Cybernetic Supermodel

Considering the complexity of the cipher context in terms ofvarious aspects designing a cipher algorithm should besystematic comprehensive and stage To design cipher al-gorithms different technologies including mathematicsphysics biometric biology and social engineering are used[21 22] Also concepts and basic sciences such as theory ofnumbers Boolean functions [23] and random functions[24 25] are very essential Depending of the application ofcipher algorithms various technical and nontechnical re-quirements should be considered for their design Detectingnew cryptography challenges such as postquantum cipherand its agility [26] and mobile applications [6 7] making analgorithm robust and the role of implementation methodsto achieve the above goals are the issues that should beconsidered in a comprehensive model Amidst consideringthe large number of components in the design of cipheralgorithms and their relationship and impact on each otherthe design and implementation of these algorithms hasbecome complicated One of the best tools to design acomplicated system is to present a model for that systeme steps associated with the design and implementationprocedure of the cipher algorithm regardless of the tripleclassification of the hash symmetric and asymmetricfunctions at the highest level are shown in Figure 1

Cryptography is one of the main information securitycomponents to transmit information from the sender to thereceiver using the most secure method [27] Design of thealgorithms has different requirements depending on itsapplication in the embedded or nonembedded system [28]To design a robust algorithm various technical and non-technical factors should be considered so that the designedalgorithm has sufficient robustness [29] On the contrary theeffective factors should be in a coherent model with logicalintegration so that their impact on each other can bemeasured and evaluated for instance in [30ndash33] variousalgorithms have been evaluated in terms of some parame-ters In fact designing a conceptual model for the cipheralgorithm requires considering all factors components and

indices that affect the design and implementation of thecipher algorithms Accordingly in the design of the cipheralgorithm there should be a balance between ldquoefficiencyrdquoand ldquoresourcesrdquo required for a specific security level [34]Considering the design and generation process of the cipheralgorithm and classification of factors components andindices the conceptual cybernetic supermodel is used fordesign and implementation Cybernetic is mainly focused onsystem performance and how they control their activitiesand communicate with their components erefore thecybernetic pattern might be a scientific basis for making thecipher algorithms targeted e cybernetic model of thecipher algorithms has four components of approachstrat-egy main process support process and control process einteractions of themain and support processes constitute thestructure of the cipher system ese interactions result in acomplicated diagram To overcome this complexity a lev-eled structure and mathematical facilities such as graph andmatrix are used Accordingly the general cybernetic modelfor the design and implementation of the cipher algorithmsis shown in Figure 2 is model is comprised of four sec-tions development approachstrategy process main processsupport process and control process e main process in-cludes cryptography algorithms e support process is di-vided into two general classes of hardware and software econtrol process includes controlling the design imple-mentation and controlling the outcome As mentioned thismodel is designed in the general level and its processes andcomponents are studied in detail in Section 3

Since the cipher algorithms have specific complexitiesthe component model is used to facilitate the processesis type of design prevents spending time on unnecessarydetails e strategic model for design of cipher algorithmsshould be presented at a level of detail that creates a trade-odd between ldquoinclusionrdquo and ldquoapplicabilityrdquo ldquoInclusionrdquoindicated including various cipher algorithms Accord-ingly considering the component extracted for the mainsupport and control section a cybernetic model can beused to design and implement cipher algorithms in threelevels as shown in Figure 3 According to this model thereis a bidirectional relationship between the main design andimplementation process and the support process at eachstep as a result of this relationship the support require-ments are demanded by the main process and provided bythe support process

31 Data Matrix of the Design Model Components Sincethere are a large number of extracted objectives or measuresin the design of cipher algorithms and some of them overlapor eliminating some of them causes no problem forachieving themain goals the criterion reductionmethod canbe used to eliminate some measures Accordingly in thisstudy the approach presented in [19] is used to reduce thenumber of measures In the feature reduction process ifeliminating one measure does not change the effective set ofthe problem it is unnecessary erefore after feature re-duction the component extraction process is carried out Inthis step three general classes of measures are extracted

Computational Intelligence and Neuroscience 3

including design and implementation objectives of cipher al-gorithms applications of cipher algorithms and imple-mentation methods of the cipher algorithms In the followingthe reduced measures are classified into the above classes and

modeling is carried out based on available measures of thesethree classes Accordingly based on the level-3 cyberneticmodel (Figure 3) the following three matrices are constitutedto design and implement the cipher algorithms

Table 1 e importance of information security needs in telemedicine technology in the field of information transmission

Data transfer Veryimportant Important No

idea Nonsignificant

Implement network protocols to ensure the transmission of information and checkits integrity 265 430 43 0

Establish a communication protocol to share information between local healthinstitutions 50 37 13 0

Encrypt important files and information 262 635 0 22Investigation of encryption mechanism by technical team of security assessor 556 328 152 0Use combinations of numbers and letters for encryption 268 25 68 0Use uppercase and lowercase letters for encryption to access remote networknetworks 50 24 174 87

Methods for controlling the integrity of application information 353 635 89 22

Hash Function

AsymmetricAlgorithms

SymmetricAlgorithms

ImplementationBasicApplication

Ciph

er T

ext

EvaluationPlai

n te

xt

Figure 1 Main process of design and implementation of cipher algorithms at the highest level

Cipher textRequest Plain text

Main Processes

Control Processes

Process

Product

Supportive Processes

Software

Hardware

CryptographyAlgorithms Policy

StrategyDevelopment

Figure 2 General schematic of the conceptual model of the cipher algorithm design

4 Computational Intelligence and Neuroscience

(i) e relationship matrix of support indices withmain design and implementation processes of thecipher algorithms since there are 13 support indicesin the level-3 model (Figure 3) and 4 steps in thelevel-1 model (Figure 2) a 13lowast4 matrix is consti-tuted to determine the relationship between themembers of these two processes where its rows arethe elements of the support process and its columnsare the quadruple elements of the design andimplementation of cryptography algorithms eelements of this matrix are between 0 and 10 asshown in Table 2 [35] e value of each elementrepresents the effectiveness of each support index oneach design and implementation step Analysis ofthis matrix and its modeling computation providesthe possibility for the policy-makers and developersto manage the resources required for each step ofdesign and implementation of the cryptographyalgorithms and use the available hardware andsoftware resources optimally

(ii) e relationship matrix of the objectives with themain design process of the cryptography algo-rithms considering the seven objectives extracted in

Section 2 and four steps in the main process todetermine the role of each step in achieving theseven objectives a 7lowast4 matrix is constituted Sincethe number of final resources used to extract data is814 papers technical reports and documents theelements of this matrix are between 0 and 814 asgiven in Table 2 In this matrix the value of eachelement represents the number of studies docu-ments and reports indicating the relationship be-tween two components An interesting point in thismatrix is that all algorithms implemented in thestudied documents are evaluated

(iii) e relationship matrix of the implementationtechniques and the design objectives of the cryp-tography algorithms considering the 29 extractedtechniques for cryptography algorithm imple-mentation and seven main objectives of the cryp-tography algorithms a 29lowast 7 matrix is constitutedto determine how much each technique is used forcryptography implementation to achieve each ob-jective (Table 3) e elements of this matrix arebetween 0 and 814 A part of this matrix is shown inTable 4

ICT

RampD

Financial

Rule amp Regulation

Standard

Human resourceEducation

OrganizationStructure

Management

Cultural

International relation

Infrastructure

Material

equipments

Hash Function

ControllerProcess

Product

Policy

Requ

est

Plai

n Te

xt

Cipher Text

Supportive

Main Processes

Software

Symmetric Algorithm

Asymmetric Algorithm

Hardware

Figure 3 e cybernetic model of design and implementation of cipher algorithms at level 3

Computational Intelligence and Neuroscience 5

311 Design of Cryptography Algorithms Using the Cyber-netic Supermodel Considering the above discussion andpresence of numerous indices and components in thedesign and implementation of the cryptography algorithmwhich make it a complex system identifying the rela-tionship between these indices and ranking them is anecessity Accordingly in this section the Q-analysis

method is used and the output of the three matrices isanalyzed According to the level three of the proposedcybernetic model the design and implementation supportprocess of the cryptography algorithms includes 13 com-ponents On the contrary the design and implementationsteps of the cryptography algorithms also include foursteps constituting a 13 lowast 4 matrix

Table 2 e 13-component interaction matrix of the support and the main design and implementation processes of cryptographyalgorithms

Main process Level 1Cryptography algorithms Level 2

Application eoretical basis Implementation Evaluation Level 3

Support process

Software

Culture 5 6 5 2 Organizationstructure 8 7 7 7

International and public relations 5 4 4 7 Financial resources 7 6 7 6

Human resources and education 8 10 10 9 Research and development 10 8 8 10

Rules 7 4 5 10 Management 9 7 8 7

FAVA 10 5 8 7 Standard 10 3 8 10

HardwareEquipment 8 6 7 8

Infrastructure 9 5 5 7Material 7 2 5 5

Table 3 e relationship matrix of the objectives with the design process of the cryptography algorithms

Main process objectives Application eoretical basis Implementation EvaluationSecurity 203 386 516 516Simplicity 5 23 38 38Resources 41 30 81 81Flexibility 15 28 45 45Scalability 6 11 22 22Speed 72 146 247 247Reliability 2 10 11 11

Table 4 A part of the interaction matrix between the cryptography algorithm implementation techniques and the seven objectives

Level 3 Security Simplicity Using resources Flexibility Scalability Speed Reliability

Support process

Avalanche effect 32 0 0 0 0 0 0Digital signature 1 0 0 0 0 0 0

Block size 1 1 0 0 2 0 0Image sharing 0 0 0 1 0 0 0

Parallel processing 0 0 0 1 12 0 0reshold technique 3 1 0 0 0 0 0

Data mining 1 0 0 0 0 0 0Binary tree 1 0 0 0 3 0 0

Cycle 5 2 0 0 12 0 0Multistage crypto 1 0 0 0 0 0 0Hybrid method 30 3 0 2 9 0 2

Hardware 0 0 7 0 0 0 0

6 Computational Intelligence and Neuroscience

32 Calculating the Incidence Matrix First the incidencematric is obtained based on the data matrix 4ndash1 is matrixrepresents the ldquoimpact of support indices on the main designprocess of the cryptography algorithmsrdquo e data matrix iscomprised of two sets D support indices set C and qua-druple design and implementation steps (Table 5) e in-cidence matrix calculated using the data matrix for α 70is represented in Table 6 By assigning different values to theparameter α difference incidence matrices are obtainederesults of the Q-analysis using C++ coding for α 70 aregiven in Figure 4

D d1 d2 d131113864 1113865

C c1 c2 c3 c41113864 1113865(1)

321 Geometric Representation Multidimensional proper-ties of the system are defined by a simple or complex setKD(C λ) such that the entities of the set D represent thesupport indices and entities of the set C represent thequadruple design and implementation steps of the cryp-tography algorithms

In the sample with α70 7 dis is as follows

d1

d2 c1 c2 c3 c41113864 1113865

d3 c41113864 1113865

d4 c1 c31113864 1113865

d5 c1 c2 c3 c41113864 1113865

d6 c1 c2 c3 c41113864 1113865

d7 c1 c41113864 1113865

d8 c1 c2 c3 c41113864 1113865

d9 c1 c3 c41113864 1113865

d10 c1 c3 c41113864 1113865

d11 c1 c3 c41113864 1113865

d12 c1 c41113864 1113865

(2)

e simplexes of σp(di) are also

σminus1 d1( 1113857 σ3 d2( 1113857 σ0 d3( 1113857 σ1 d4( 1113857 σ3 d5( 1113857 σ0 d3( 1113857 σ1 d7( 1113857

σ3 d8( 1113857 σ2 d9( 1113857 σ2 d10( 1113857 σ2 d11( 1113857 σ1 d12( 1113857 σ1 d13( 1113857 (3)

erefore the complex dimension is 3 In other wordsthe diagnosis classes d2 (structureorganization) d5 (humanresourceseducation) d6 (research and development) andd8 (management) have the largest dimension

33 CalculatingDimensions andQ-Link Q-link is defined asthe link between a subset with smallest interface between twosubsequent dis in the chain of d1 to dn Q-link between twosubsequent dis with α70 7 is

Table 5 Sets of dis and cis support indices based on the datamatrix

d1 Culture

d2 Structureorganizationd3 Internationalpublic relationsd4 Financiald5 Human resourceseducationd6 Research and developmentd7 Rulesd8 Managementd9 FAVAd10 Standardd11 Equipmentd12 Infrastructured13 MaterialC1 ApplicationC2 eoretical basisC3 ImplementationC4 Evaluation

Table 6 e incidence matrix of the support indicesrsquo impact of thedesign steps of the cryptography algorithms with α 70

C1 C2 C3 C4

d1 0 0 0 0d2 1 1 1 1d3 0 0 0 1d4 1 0 1 0d5 1 1 1 1d6 1 1 1 1d7 1 0 0 1d8 1 1 1 1d9 1 0 1 1d10 1 0 1 1d11 1 0 1 1d12 1 0 0 1d13 1 0 0 0

Computational Intelligence and Neuroscience 7

σminus1 d1( 1113857 σ3 d2( 1113857⟶ 1 σ3 d2( 1113857 σ0 d3( 1113857⟶ 0 σ0 d3( 1113857 σ1 d4( 1113857⟶ minus 1

σ1 d4( 1113857 σ3 d5( 1113857⟶ 1 σ3 d5( 1113857 σ3 d6( 1113857⟶ 3 σ3 d6( 1113857 σ1 d7( 1113857⟶ 1

σ1 d7( 1113857 σ3 d8( 1113857⟶ 1 σ3 d8( 1113857 σ2 d9( 1113857⟶ 2 σ2 d9( 1113857 σ2 d10( 1113857⟶ 2

σ2 d10( 1113857 σ2 d11( 1113857⟶ 2 σ3 d11( 1113857 σ1 d12( 1113857⟶ 1 σ1 d12( 1113857 σ0 d13( 1113857⟶ 0

(4)

e maximum link dimension is 3 indicating the re-lationship between the diagnosis classes

331 Calculating the Structure Vectors As mentioned inthe definitions the vector Qq is a simplification basiscreated to eliminate the additional impacts in the equiv-alent simplex sets e maximum complex dimension withα70 7 is 3 erefore the first structure vector based onthe output is

Dimension 3 2 1 0 (5)

e second structure vector P is

dimensions 3 2 1 0

P Pdim3 Pdim2 Pdim1 Pdim0( 1113857

P 4 7 10 12( 1113857

(6)

where Pq is the number simplexes greater than or equal to qin the set K in which P is the number of simplex linkrepetitions (support indices) in the quadruple design andimplementation steps of the cryptography algorithms Basedon the values of these two vectors it is seen that the rela-tionship between the support indices and the quadrupledesign and implementation steps of the cryptography al-gorithms is high is issue indicates the role of supportcomponents in the design and implementation of the

cryptography algorithms which should be consideredseriously

332 Calculating the Obstruction or Flexibility VectorQlowastK represents the number of structural obstructions forsimplex interactions in dimension k

Qlowast

Q minus I⟶ Qlowast

1 1 1 11113858 1113859 minus 1 1 1 11113858 1113859⟶ Qlowast

0 0 0 01113858 1113859

(7)

As can be seen there is no obstruction in any of thecommunication levels indicating that there is a significantrelationship between the support components at eachequivalence class

333 Calculating Irregularity e value of (eccprime(σ)) iscalculated using the Chinese method e results for α70

7 are shown in Table 7 e calculated irregularity valueshows that the indices d2 (structureorganization) d5 (hu-man resourceseducation) d6 (research and development)and d8 (management) affect other indices

334 Calculating Complexity Also the results ofQ-analysiscan be used to describe structure complexity According toequations (4)ndash(9) and for α = 7 the complexity measure is

-1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1

3 0 1 3 3 1 3 2 2 2 1

0 -1 0 0 0 0 0 0 0 0

1 1 1 0 1 1 1 1 0

3 3 1 3 2 2 2 1

3 1 3 2 2 2 1

3 2 2 2 1

2 2 2 1

2 2 1

2 1

1

1 1 1 1 1 1

-1

-1

0

0

0

0

0

0

0

0

0

0

0q Q3210

11

11

SETSX2 X5 X6 X8X2 X5 X6 X8 X9 X10 X11X2 X4 X5 X6 X7 X8 X9 X10 X11 X12X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13

Figure 4 Implementation results of the model for support components of the cryptography algorithmsrsquo design with α 70

8 Computational Intelligence and Neuroscience

Q 1 1 1 1( 1113857

ψ(K) 2(1 + 2 + 3 + 4)

(4lowast 5)1113890 1113891 1

(8)

Since in the above model there is no obstruction amongthe component of the equivalent class it was expected thatthe complexity of the support components is not high andthe obtained complexity index of 1 verifies this expectatione system complexity for different alpha-cuts is shown inFigure 5

335 Ranking the Support Components of the Design andImplementation of the Cryptography Algorithms e resultsof using A-analysis are shown in Table 7 e connectionstrength of the factors in one group is specified with alpha-cut erefore the support components are grouped in 5levels Each level describes the priority and importance of thegroup in developing the cryptography algorithms In Fig-ure 6 the ranking pyramid of the support components usingQ-analysis is shown To allocate proper resources thecomponents existing in higher levels of the pyramid (Fig-ure 6) are of higher priority

336 Validation of the Results In this section the resultsof the cybernetic model and Q-analysis for supportcomponentsrsquo ranking are compared with the results re-ported in the global cybersecurity index (GCI) in 20152017 and 2018 presented by ITU [36ndash39] e GCI reportsare focused on five indices including ldquolegal cases orga-nization necessities technical issues capacity buildingand cooperationrdquo and the subindices include legaltechnical organization capacity building and coopera-tion According to the presented indices and subindices itis clear that the ldquorulesrdquo ldquostandardrdquo ldquoresearch and de-velopmentrdquo ldquoeducationrdquo and ldquomanagementrdquo are ofhigher priority in security establishment Although ourresearch is more skilled and detailed compared to the GCIreports but the results verify our findings Also Table 8shows the ranking of the support components obtainedusing Q-analysis

337 Executing the Model and Analyzing the Results of theObjectivesrsquo Impact on Design Steps of the Cryptography Al-gorithmsrsquo Matrix In this section the role of the sevencomponents on the quadruple design and implementationsteps of the cryptography algorithm is analyzed with Q-analysis Using the Q-analysis method and the 7lowast4 matrixobtained from the relationship of the cryptography algo-rithmsrsquo design objectives on their quadruple steps theirindices are ranked

338 Calculating the Incidence Matrix and Executing theModel First the incidence matrix is obtained for the datamatrix shown in Figure 2 It can be seen in Table 9 that eachelement of the two sets represents which indice e inci-dence matrix calculated from the data matrix for α5 40 is

given in Table 10 e results of implementing the model forα5 40 are given in Figure 7

4 Geometric Representation

In the sample with α5 40 dis are

d1 c1 c2 c3 c41113864 1113865

d2

d3 c1 c3 c41113864 1113865

d4 c3 c41113864 1113865

d5

d6 c1 c2 c3 c41113864 1113865

d7

(9)

e simplexes of σp(di) are also

σ3 d1( 1113857σminus1 d2( 1113857σ2 d3( 1113857σ1 d4( 1113857σminus1 d5( 1113857σ3 d6( 1113857σminus1 d7( 1113857

(10)

erefore the complex dimension is 3 In other wordsthe discriminant classes d1 (security) and d6 (speed) have thelargest dimension

41 Calculating the Structure Vectors As mentioned thevector Qq is a simplification basis to eliminate the additionaleffects in the set of equivalent simplexes e maximumcomplex dimension for α5 40 is 3 erefore the firststructure vector based on the output is

Dimension 3 2 1 1

Q 1 1 1 1( 1113857

Dimension 3 2 1 0

P 2 3 4 7( 1113857

(11)

e second structure vector P is

P 2 3 4 7( 1113857 (12)

42 Calculating the Obstruction Vector or InflexibilityQlowastK represents the number of structural obstructions forsimplex interactions in dimension k which is as follows forα5 40

1 1 1 1 1332

005

115

225

com

plex

ity

alpha cutα=5 α=6 α=7 α=8 α=9 α=10

Figure 5 System complexity of the support indices of design andimplementation of the cryptography algorithms for different alpha-cuts

Computational Intelligence and Neuroscience 9

Qlowast

Q minus I⟶ Qlowast

2 1 1 31113858 1113859 minus 1 1 1 11113858 1113859⟶ Qlowast

0 0 0 01113858 1113859

(13)

According to the obtained values it can be concludedthat there are no structural obstructions among the main

design indices of the cryptography algorithms at ismultiple objectives are considered simultaneously by thecryptography algorithm designers

Table 7 e irregularity of the data matrix parameters for α= 7

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d2 d5 d6 d8 qi 3 2 1 364 3 061d9 d10d11 qi 2 1 064 3 011d4 d7 d12 qi 1 014 3 002

human resourceRampD

rules FAVA standard

management infrastructure

organizationstructure equipmentinternational and public relations finanacial

resources and material

Figure 6 e ranking pyramid of the support components using Q-analysis

Table 8 Ranking of the support components using Q-analysis

Relationship of the support components with the design and implementation of the cryptography algorithms (q 0)No relationship α 0 complete relationship α 100

Human resource RampD rules FAVA and standard α100 10Management and infrastructure α90 9Organizationstructure and equipment α80 8International and public relations financial resources and material α70 7Culture α60 6All components α60 5

Table 9 Set of dis and cis the objectivesrsquo indices

eoretical basis C2 Reliability d7 Scalability d5 Resources d3 Security d1Implementation C3 Application C1 Speed d6 Flexibility d4 Simplicity d2Evaluation C4

Table 10 e incidence matrix of the design objectives of thecryptography algorithms with α= 40

c1 c2 c3 c4

d1 1 1 1 1d2 0 0 0 0d3 1 0 1 1d4 0 0 1 1d5 0 0 0 0d6 1 1 1 1d7 0 0 0 0

-1 -1 -1 -1 -1

2 1 -1 2

1 -1 1

-1 -1

3

3 -1 2 1 -1 3 -1

-1

-1

-1

-1

-1

-1q Q3210

11

11

SETSX1X6X1X3X6X1X3X4X6X1X3X4X6

Figure 7 e results of executing the model for design objectivesrsquoindices with α 40

10 Computational Intelligence and Neuroscience

43 Calculating Irregularity e results of applying theChinese method (ecclsquo(σ)) for calculating irregularity forα5 40 are shown in Table 11 e calculated irregu-larity value shows that indices d1 (security) and d6(speed) have received more attention compared to otherindices

44 Calculating Complexity Structural complexity for asample with α5 40 is

dimensions 3 2 1 0

Q Qdim3 Qdim2 Qdim1 Qdim0( 1113857

Q 1 1 1 1( 1113857

ψ(K) 2(1 + 2 + 3 + 4)

(4lowast 5)1113890 1113891 1

(14)

Since in the above model there was no obstructionamong the components of the equivalent classes in any of thecommunication levels it was expected that there is not ahigh complexity among objectives of the cryptography al-gorithm design and the complexity index of 1 verified thisexpectation e system complexity for different alpha-cutsis shown in Figure 8 As can be seen the system complexity is1 for all significant values of alpha

45 Prioritizing the Parameters of the Main Objectives of theCryptography Algorithm Design Using Q-AnalysisAccording to the analysis results the parameters can beclassified into 5 levelsese levels are shown in Table 12 andFigure 9 Each level indicates a priority and importance ofthe group in the development of cryptography algorithmsTo allocate proper resources the components at the higherlevels of the pyramid are of higher priority As can be seenthree objectives of security speed and optimal usage ofresources have the highest priority for the design of cryp-tography algorithms

46 Executing the Model and Analyzing the Results for theCryptography Algorithm Implementation Techniques einteraction matrix between the 29 extracted techniques andthe sevenmain objectives is shown in Figure 10e purposeof this section is to rank the cryptography algorithmimplementation techniques to achieve the goals of interest

47Calculating the IncidenceMatrixandExecuting theModele data matrix A is comprised of two sets e set Drepresents the employed techniques and the set C representsthe seven main design objectives (Table 13)

D d1 d2 d291113864 1113865

C c1 c2 c71113864 1113865(15)

Some parts of the incidence matrix calculated from thedata matrix A for α3 are shown in Table 14 and the resultsof theQ-analysis model for α3 24 are shown in Figure 11

48 Geometric Representation In the sample with α3 24the dis are

d1 c11113864 1113865

d2

d3

d4

d5

d6

d7

d8

d9

d10

d11 c11113864 1113865

d12

d13 c51113864 1113865

d14

d15 c1 c4 c51113864 1113865

d16

d17

d18

d19 c11113864 1113865

d20

d21

d22 c31113864 1113865

d23

d24

d25

d26

d27 c11113864 1113865

d28

d29

(16)

e simplexes of σp(di) are

Computational Intelligence and Neuroscience 11

σ0 d1( 1113857

σminus1 d2( 1113857

σminus1 d3( 1113857

σminus1 d4( 1113857

σminus1 d5( 1113857

σminus1 d6( 1113857

σminus1 d7( 1113857

σminus1 d8( 1113857

σminus1 d9( 1113857

σminus1 d10( 1113857

σ0 d11( 1113857

σminus1 d12( 1113857

σ0 d13( 1113857

σminus1 d14( 1113857

σ2 d15( 1113857

σminus1 d16( 1113857

σminus1 d17( 1113857

σminus1 d18( 1113857

σ0 d19( 1113857

σminus1 d20( 1113857

σminus1 d21( 1113857

σ0 d22( 1113857

σminus1 d23( 1113857

σminus1 d24( 1113857

σminus1 d25( 1113857

σminus1 d26( 1113857

σ0 d27( 1113857

σminus1 d28( 1113857

σminus1 d29( 1113857

(17)

erefore the complex dimension is 2 In other wordsthe discriminant class d15 (basic sciences) has the largestdimension

49 Calculating the Dimensions and the Q-Link In the alphadefined for cryptography algorithm implementation tech-niques to achieve the defined goals the obtained q-linkshows that these techniques are relatively independent al-though there is a weak relationship between some tech-niques Q-link in the samples with α3 24 between eachtwo subsequent di is as follows

σ0 d1( 1113857 σminus1 d2( 1113857⟶ minus 1

σminus1 d2( 1113857 σminus1 d3( 1113857⟶ minus 1

σ0 d3( 1113857 σminus1 d4( 1113857⟶ minus 1

σminus1 d4( 1113857 σminus1 d5( 1113857⟶ minus 1

σminus1 d5( 1113857 σminus1 d6( 1113857⟶ minus 1

σminus1 d6( 1113857 σminus1 d7( 1113857⟶ minus 1

σminus1 d7( 1113857 σminus1 d8( 1113857⟶ minus 1

σminus1 d8( 1113857 σminus1 d9( 1113857⟶ minus 1

σminus1 d9( 1113857 σminus1 d10( 1113857⟶ minus 1

σminus1 d10( 1113857 σ0 d11( 1113857⟶ minus 1

σ0 d11( 1113857 σminus1 d12( 1113857⟶ minus 1

σminus1 d12( 1113857 σ0 d13( 1113857⟶ minus 1

σ0 d13( 1113857 σminus1 d14( 1113857⟶ minus 1

σminus1 d14( 1113857 σ2 d15( 1113857⟶ minus 1

σ2 d15( 1113857 σminus1 d16( 1113857⟶ minus 1

σminus1 d16( 1113857 σminus1 d17( 1113857⟶ minus 1

σminus1 d17( 1113857 σminus1 d18( 1113857⟶ minus 1

σminus1 d18( 1113857 σ0 d19( 1113857⟶ minus 1

σ0 d19( 1113857 σminus1 d20( 1113857⟶ minus 1

σminus1 d20( 1113857 σminus1 d21( 1113857⟶ minus 1

σminus1 d21( 1113857 σ0 d22( 1113857⟶ minus 1

σ0 d22( 1113857 σminus1 d23( 1113857⟶ minus 1

σminus1 d23( 1113857 σminus1 d24( 1113857⟶ minus 1

σminus1 d24( 1113857 σminus1 d25( 1113857⟶ minus 1

σminus1 d25( 1113857 σminus1 d26( 1113857⟶ minus 1

σminus1 d26( 1113857 σ0 d27( 1113857⟶ minus 1

σ0 d27( 1113857 σminus1 d28( 1113857⟶ minus 1

σminus1 d28( 1113857 σminus1 d29( 1113857⟶ minus 1

(18)

Table 11 Irregularity of the cryptography objectivesrsquo parameters in the data matrix A for α5

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d1 and d6 qi 3 2 and 1 433 3 072d3 qi 2 and 1 133 3 026d4 qi 1 033 3 006

11117

1 1 1

09095

1105

11115

12

com

plex

ity

alpha cutα=2 α=3 α=5 α=15 α=25

Figure 8 System complexity of the cryptography algorithm designobjectives for different alpha-cuts

12 Computational Intelligence and Neuroscience

410 Calculating the Structure Vectors erefore the firststructure vector based on the software output is

Q 1 1 2( 1113857 (19)

e second structure vector P is

P 1 1 29( 1113857 (20)

Since the large values of P of higher dimensions dem-onstrate more links the second structure vector calculatedfor the alpha of interest shows that the relationship betweenthe cryptography algorithm implementation techniques isminimum

411 Calculating the Obstruction or Inflexibility Vectore obstruction vector (Qlowast) for a sample with α3 24 is

Table 12 Ranking the objectivesrsquo parameters considering different cuts in Q-analysis

e equivalence class for the parameters with minimum relation level (q 0) αSecurity α50 244Speed α30 244Resources α15 122Flexibility α5 40Simplicity-scalability α2 16(All parameters)

securityspeed

optimal usage of resources

flexibility

simplicity-scalability

Figure 9 Prioritizing the design and implementation objectives of the cryptography algorithms using Q-analysis

securityspeed

optimal usage of resources

flexibility

simplicity-scalability-reliability

Figure 10 Prioritizing the design and implementation objectives of the cryptography algorithms using Q-analysis

Table 13 Set of dis and cis the techniques employed for implementing cryptography algorithms

Flexibility C4 Fuzzy logic d25 Storage space d17 Cycle d9 Avalanche effect d1Scalability C5 Software d26 Clustering d18 Multiple step d10 Digital signature d2Speed C6 Steganography d27 Key d19 Hybrid method d11 Block size d3Reliability C7 Music harmony d28 Graph d20 Hardware d12 Image sharing d4

Artificial intelligence d29 Characteristic oriented d21 Hardware d13 Parallel processing d5Security C1 Energy consumption d22 Occupation area d14 reshold technique d6Simplicity C2 Bandwidth consumption d23 Basic science d15 Data mining d7Resources C3 Memory consumption d24 Compression d16 Binary tree d8

Table 14 e incidence matrix of the cryptography algorithmimplementation techniques with α 24

C1 C2 C3 C4 C5 C6 C7

d1 1 0 0 0 0 0 0d2 0 0 0 0 0 0 0d3 0 0 0 0 0 0 0d4 0 0 0 0 0 0 0d5 0 0 0 0 0 0 0d6 0 0 0 0 0 0 0d7 0 0 0 0 0 0 0d8 0 0 0 0 0 0 0d9 0 0 0 0 0 0 0d10 0 0 0 0 0 0 0d11 1 0 0 0 0 0 0d12 0 0 0 0 0 0 0

Computational Intelligence and Neuroscience 13

Qlowast

Q minus I⟶ Qlowast

1 1 21113858 1113859 minus 1 1 11113858 1113859⟶ Qlowast

0 0 11113858 1113859

(21)

e value of QlowastK represents the number of structurallimitations or obstructions for the cryptography techniquesrsquointeraction at dimension k As can be seen from the cal-culated obstruction vector there is a significant relationshipat some levels

412 Calculating Irregularity Irregularity is the integrationdegree of a cryptography method in the total complex Mea-suring irregularity (eccrsquo) for α3 24 is shown in Table 15 Ascan be seen irregularity for the discriminant class d15 indicatesthat this technique is isolated from other techniques

413 Calculating Complexity Structure complexity for asample with α3 24 is

Q 1 1 2( 1113857

ψ(K) 2(2 + 2 + 3)

(3lowast 4)1113890 1113891 117

(22)

e system complexity for different values of alpha isshown in Figure 12 As can be seen the total complexity ofthe system for large values alpha tends to stability

414 Ranking the Cryptography Algorithm ImplementationTechniques ese techniques can be classified into 5 levelsusing the analysis of the obtained results ese levels can beseen in Table 16 and Figure 13 e link power of the factorsin one group is specified with alpha Each level describespriority and importance of the group in the development ofcryptography algorithms Accordingly six methods of basicscience key management using hardware methods usingsteganography Avalanche effect and using hybrid method

q Q21o

112

SETSX15X15X1X11X15X19X27X13 amp X22

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-10-1-1-10-1-1-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-10-1-1-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

2-10-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1

-1-1-1-1-1-1

-1-1-1-1-1

-1-1-1-1

-1-1-1

-1-10

Figure 11 Results of executing the model for cryptography algorithm implementation techniques with α3 24

14 Computational Intelligence and Neuroscience

are the most important cryptography algorithm imple-mentation techniques to achieve the main objectives

5 Conclusion

In this study the design and implementation model of thecryptography algorithms is designed in three levels and itseffective components are extracted To organize the com-ponents in addition to elimination of the additional com-ponents through the measure reduction algorithm a properclassification is applied to examine the mutual effects Afterclassification three 13lowast 4 7lowast 4 and 29lowast 7 matrices areconstituted and the model is implemented on these threematrices To implement the designed model Q-analysis isused Accordingly for support indices five indices of highpriority include human resources research and develop-ment management organizationstructure and equipmentFor the algorithm design objectives index five high priorityindices include security speed optimal usage of resources

and simplicity For the indices related to implementationtechniques of the cryptography algorithms the most appliedtechniques for achieving the determined objectives includeusing basic science hardware and software methods usingkeymanagement hybrid method steganography Avalancheeffect and using artificial intelligence In the future work wewill consider the following cases

A plan should be formulated according to the prioritiesAccording to the outputs of this study and the presented pri-orities the following topics can be investigated in future studies

(1) Presenting a comprehensive model for generatingvarious cryptography algorithms based on the pri-orities of interest

(2) Examining and formulating a model about the roleof basic science for design and implementation ofcryptography algorithms considering the significantrole of ldquobasic sciencerdquo in implementation of thecryptography algorithms and about the role ofschools and universities

(3) Presenting a model for design and implementationof cryptography algorithms in the IoT infrastructurewith optimal resource usage

Data Availability

e data used to support the findings of the study areavailable from the corresponding author upon request

Conflicts of Interest

e authors declare that there are no conflicts of interestregarding the publication of this article

Table 15 Irregularity of the cryptography algorithm implementation techniques for α3 24

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d15 qi 2 1 3 2 1- qi 1 1 2 033

11117

1 1 1

091

1112

com

plex

ity

alpha-cutα=2 α=3 α=5 α=15 α=25

Figure 12 System complexity of cryptography algorithm implementation techniques for different alpha-cuts

Table 16 Ranking the implementation techniques of the cryptography algorithms considering various alpha-cuts using Q-analysis

Equivalence classes for the parameters with minimum relationship level (q 0) αBasic sciences α20 163Key α10 81Hardware and steganography α5 40Avalanche effect and hybrid method α3 24Artificial intelligence software occupation area energy consumption memory consumption α2 16

basic sciences

key

Avalanche effect hybrid method

hardware steganography

artificial intelligence soware occupationarea energy consumption memory consumption

Figure 13 Prioritizing the implementation techniques of thecryptography algorithms using Q-analysis

Computational Intelligence and Neuroscience 15

Acknowledgments

is work was partially supported by Iran National ScienceFoundation (INSF) under contract No 9653979

References

[1] D Nilesh and M Nagle ldquoe new cryptography algorithmwith high throughputrdquo in Proceedings of the 2014 Interna-tional Conference on Computer Communication and Infor-matics pp 3ndash7 Coimbatore India January 2014

[2] S Ravi P Kocher R Lee G McGraw and A RaghunathanldquoSecurity as a new dimension in embedded system designrdquo inProceedings of the 41st annual conference on Design auto-mation-DAC lsquo04 pp 753ndash760 San Diego CA USA March2004

[3] P Kuppuswamy and S Q Y A Khalidi ldquoHybrid encryptiondecryption technique using new public key and symmetric keyalgorithmrdquo International Journal of Information and Com-puter Security vol 6 no 4 pp 372ndash382 2014

[4] S Bhat and V Kapoor ldquoSecure and efficient data privacyauthentication and integrity schemes using hybrid cryptog-raphyrdquo International Conference on Advanced ComputingNetworking and Informatics vol 870 pp 279ndash285 2019

[5] D Ott and C Peikert ldquoAnd other workshop participantsldquoidentifying research challenges in post quantum cryptog-raphy migration and cryptographic agilityrdquo 2019 httparxivorgabs190907353

[6] P Manickam K Shankar E Perumal M Ilayaraja andK Sathesh Kumar ldquoSecure data transmission through reliablevehicles in VANET using optimal lightweight cryptographyrdquoin Advanced Sciences and Technologies for Security Applica-tions pp 193ndash204 Springer New York NY USA 2019

[7] S Prakash and A Rajput ldquoHybrid cryptography for securedata communication in wireless sensor networksrdquo Advancesin Intelligent Systems and Computing vol 696 pp 589ndash5992018

[8] A A Soofi I Riaz and U Rasheed ldquoAn enhanced vigenerecipher for data securityrdquo International Journal of Scientific ampTechnology Research vol 4 no 8 pp 141ndash145 2015

[9] N Shamsuddin M Syafiqah and S Ali Pitchay ldquoImple-menting location-based cryptography on mobile applicationdesign to secure data in cloud storagerdquo Journal of PhysicsConference Series IOP Publishing vol 1551 no 1 2020

[10] L Jiao H Yonglin and D Feng ldquoStream cipher designs areviewrdquo Science China Information Sciences vol 63 no 3pp 1ndash25 2020

[11] NIST ldquoNIST cryptographic standards and guidelines devel-opment processrdquo Inside NIST vol 27 2016

[12] G Alagic J Alperin-Sheriff D Apon et al Status Report onthe First Round of the NIST Post-Quantum CryptographyStandardization Process National Institute of Standards andTechnology Gaithersburg MD USA pp 1ndash27 2019

[13] I Damaj and S Kasbah ldquoAn analysis framework for hardwareand software implementations with applications from cryp-tographyrdquo Computers amp Electrical Engineering vol 69pp 572ndash584 2018

[14] M S Croock Z A Hassan and S D Khudhur ldquoAdaptive keygeneration algorithm based on software engineering meth-odologyrdquo International Journal of Electrical and ComputerEngineering vol 11 no 1 p 589 2021

[15] K Sha T A Yang W Wei and S Davari ldquoA survey of edgecomputing-based designs for iot securityrdquo Digital Commu-nications and Networks vol 6 no 2 pp 195ndash202 2020

[16] M Yahuza M Y I B Idris A W B A Wahab et alldquoSystematic review on security and privacy requirements inedge computing state of the art and future research oppor-tunitiesrdquo IEEE Access vol 8 pp 76541ndash76567 2020

[17] A Bhowmik J Dey and S Karforma A Novel Double TierCryptographic System (nDTCS) to Reinforce Patientsrsquo Privacyin Contemporary COVID-19 Telemedicine Springer NewYork NY USA 2021

[18] N Gupta User Integrity Protection Security Model forEnhanced Telemedicine for Healthcare Networks 2020httpsosfio69jdq

[19] J Dey A Bhowmik A Sarkar S Karforma andB Chowdhury Cryptographic Engineering on COVID-19Telemedicine An Intelligent Transmission through RecurrentRelation Based Session Key Springer New York NY USA2021

[20] V Hosseinian H Ayatollahi H Haghani and E MehraeenldquoRequirements of information security in a telemedicinenetwork review of IT managersrsquo opinionrdquo Journal of Para-medical Sciences amp Rehabilitation vol 4 no 2 pp 31ndash402015

[21] D Karaoglan Altop A Levi and V Tuzcu ldquoDeriving cryp-tographic keys from physiological signalsrdquo Pervasive andMobile Computing vol 39 pp 65ndash79 2017

[22] O Uzunkol and M S Kiraz ldquoStill wrong use of pairings incryptographyrdquo Applied Mathematics and Computationvol 333 pp 467ndash479 2018

[23] S Picek D Jakobovic J F Miller L Batina and M CupicldquoCryptographic Boolean functions one output many designcriteriardquo Applied Soft Computing vol 40 pp 635ndash653 2016

[24] R Saha and G Geetha ldquoSymmetric random function gen-erator (SRFG) a novel cryptographic primitive for designingfast and robust algorithmsrdquo Chaos Solitons amp Fractalsvol 104 pp 371ndash377 2017

[25] A Achuthshankar and A Achuthshankar ldquoA novel sym-metric cryptography algorithm for fast and secure encryp-tionrdquo in Proceedings of the 2015 IEEE 9th InternationalConference on Intelligent Systems and Control (ISCO) pp 4ndash9Coimbatore India January 2015

[26] D Ott and C Peikert ldquoIdentifying research challenges in postquantum cryptography migration and cryptographic agilityrdquo2019 httparxivorgabs190907353

[27] W Liu B Ying H Yang and H Wang ldquoAccurate modelingfor predicting cryptography overheads on wireless sensornodesrdquovol 2 pp 997ndash1001 in Proceedings of the 2019 11thInternational Conference on Advanced CommunicationTechnology vol 2 pp 997ndash1001 IEEE Bangalore IndiaFebruary 2019

[28] S Feizi A Ahmadi and A Nemati ldquoA hardware imple-mentation of simon cryptography algorithmrdquo in Proceedingsof the 2014 4th International Conference on Computer andKnowledge Engineering (ICCKE) pp 245ndash250 IEEEMashhad Iran 2014

[29] E ambiraja G Ramesh and R Umarani ldquoA survey onvarious most common encryption techniquesrdquo 2012 httpwwwijarcssecom

[30] A Gupta and N K Walia ldquoCryptography Algorithms areviewrdquo International Journal of Engineering Research andDevelopment vol 2 no 2 pp 1667ndash1672 2014

[31] R Bhanot and R Hans ldquoA review and comparative analysis ofvarious encryption algorithmsrdquo International Journal of Se-curity and Its Applications vol 9 no 4 pp 289ndash306 2015

[32] S Chandra S Paira S S Alam and G Sanyal ldquoA com-parative survey of symmetric and asymmetric key

16 Computational Intelligence and Neuroscience

cryptographyrdquo in Proceedings of the 2014 InternationalConference on Electronics Communication and Computa-tional Engineering (ICECCE) pp 83ndash93 Hosur India No-vember 2014

[33] F Maqsood M Ahmed M Mumtaz and M Ali ldquoCryp-tography a comparative analysis for modern techniquesrdquoInternational Journal of Advanced Computer Science andApplications vol 8 no 6 p 6 2017

[34] S B Sadkhan and A O Salman ldquoA survey on lightweight-cryptography status and future challengesrdquo in Proceedings ofthe International Conference on Advances in SustainableEngineering and Applications pp 105ndash108 Wasit Iraq June2018

[35] N V oai ldquoCriteria and dimension reduction of linearmultiple criteria optimization problemsrdquo Journal of GlobalOptimization vol 52 no 3 pp 499ndash508 2012

[36] C E Shannon ldquoCommunication theory of secrecy systems1945rdquo MD Computing vol 15 no 1 pp 57ndash64 1998

[37] L J Fennelly M Beaudry and M A Perry Security in 2025[38] ITU Global Cybersecurity Index 2018 2019[39] ITU Global Cybersecurity Index (GCI) 2017 2017

Computational Intelligence and Neuroscience 17

purpose matrix confusion operations are used e inno-vation of this solution is in the process of matrix transferwhich is transmitted in the form of a number of cardiologistsin particular Finally the efficiency of the proposed methodis evaluated with regard to cryptographic engineeringtransparency and strength e results indicate that thismethod provides more security in the medical data trans-mission process Hosseinian et al [20] examined the im-portance of information security needs in telemedicinetechnology in the field of information transmission In thisstudy the data collection tool was a questionnaire that wasdesigned based on the criteria of the Association of Infor-mation Management and Health Care Systems (HMISS) inthe field of telemedicine network security and securitystandards of the American Telemedicine Association isquestionnaire has been calculated separately based on ascore of 1 to 5 Table 1 summarizes the importance of eachsection

3 Cybernetic Supermodel

Considering the complexity of the cipher context in terms ofvarious aspects designing a cipher algorithm should besystematic comprehensive and stage To design cipher al-gorithms different technologies including mathematicsphysics biometric biology and social engineering are used[21 22] Also concepts and basic sciences such as theory ofnumbers Boolean functions [23] and random functions[24 25] are very essential Depending of the application ofcipher algorithms various technical and nontechnical re-quirements should be considered for their design Detectingnew cryptography challenges such as postquantum cipherand its agility [26] and mobile applications [6 7] making analgorithm robust and the role of implementation methodsto achieve the above goals are the issues that should beconsidered in a comprehensive model Amidst consideringthe large number of components in the design of cipheralgorithms and their relationship and impact on each otherthe design and implementation of these algorithms hasbecome complicated One of the best tools to design acomplicated system is to present a model for that systeme steps associated with the design and implementationprocedure of the cipher algorithm regardless of the tripleclassification of the hash symmetric and asymmetricfunctions at the highest level are shown in Figure 1

Cryptography is one of the main information securitycomponents to transmit information from the sender to thereceiver using the most secure method [27] Design of thealgorithms has different requirements depending on itsapplication in the embedded or nonembedded system [28]To design a robust algorithm various technical and non-technical factors should be considered so that the designedalgorithm has sufficient robustness [29] On the contrary theeffective factors should be in a coherent model with logicalintegration so that their impact on each other can bemeasured and evaluated for instance in [30ndash33] variousalgorithms have been evaluated in terms of some parame-ters In fact designing a conceptual model for the cipheralgorithm requires considering all factors components and

indices that affect the design and implementation of thecipher algorithms Accordingly in the design of the cipheralgorithm there should be a balance between ldquoefficiencyrdquoand ldquoresourcesrdquo required for a specific security level [34]Considering the design and generation process of the cipheralgorithm and classification of factors components andindices the conceptual cybernetic supermodel is used fordesign and implementation Cybernetic is mainly focused onsystem performance and how they control their activitiesand communicate with their components erefore thecybernetic pattern might be a scientific basis for making thecipher algorithms targeted e cybernetic model of thecipher algorithms has four components of approachstrat-egy main process support process and control process einteractions of themain and support processes constitute thestructure of the cipher system ese interactions result in acomplicated diagram To overcome this complexity a lev-eled structure and mathematical facilities such as graph andmatrix are used Accordingly the general cybernetic modelfor the design and implementation of the cipher algorithmsis shown in Figure 2 is model is comprised of four sec-tions development approachstrategy process main processsupport process and control process e main process in-cludes cryptography algorithms e support process is di-vided into two general classes of hardware and software econtrol process includes controlling the design imple-mentation and controlling the outcome As mentioned thismodel is designed in the general level and its processes andcomponents are studied in detail in Section 3

Since the cipher algorithms have specific complexitiesthe component model is used to facilitate the processesis type of design prevents spending time on unnecessarydetails e strategic model for design of cipher algorithmsshould be presented at a level of detail that creates a trade-odd between ldquoinclusionrdquo and ldquoapplicabilityrdquo ldquoInclusionrdquoindicated including various cipher algorithms Accord-ingly considering the component extracted for the mainsupport and control section a cybernetic model can beused to design and implement cipher algorithms in threelevels as shown in Figure 3 According to this model thereis a bidirectional relationship between the main design andimplementation process and the support process at eachstep as a result of this relationship the support require-ments are demanded by the main process and provided bythe support process

31 Data Matrix of the Design Model Components Sincethere are a large number of extracted objectives or measuresin the design of cipher algorithms and some of them overlapor eliminating some of them causes no problem forachieving themain goals the criterion reductionmethod canbe used to eliminate some measures Accordingly in thisstudy the approach presented in [19] is used to reduce thenumber of measures In the feature reduction process ifeliminating one measure does not change the effective set ofthe problem it is unnecessary erefore after feature re-duction the component extraction process is carried out Inthis step three general classes of measures are extracted

Computational Intelligence and Neuroscience 3

including design and implementation objectives of cipher al-gorithms applications of cipher algorithms and imple-mentation methods of the cipher algorithms In the followingthe reduced measures are classified into the above classes and

modeling is carried out based on available measures of thesethree classes Accordingly based on the level-3 cyberneticmodel (Figure 3) the following three matrices are constitutedto design and implement the cipher algorithms

Table 1 e importance of information security needs in telemedicine technology in the field of information transmission

Data transfer Veryimportant Important No

idea Nonsignificant

Implement network protocols to ensure the transmission of information and checkits integrity 265 430 43 0

Establish a communication protocol to share information between local healthinstitutions 50 37 13 0

Encrypt important files and information 262 635 0 22Investigation of encryption mechanism by technical team of security assessor 556 328 152 0Use combinations of numbers and letters for encryption 268 25 68 0Use uppercase and lowercase letters for encryption to access remote networknetworks 50 24 174 87

Methods for controlling the integrity of application information 353 635 89 22

Hash Function

AsymmetricAlgorithms

SymmetricAlgorithms

ImplementationBasicApplication

Ciph

er T

ext

EvaluationPlai

n te

xt

Figure 1 Main process of design and implementation of cipher algorithms at the highest level

Cipher textRequest Plain text

Main Processes

Control Processes

Process

Product

Supportive Processes

Software

Hardware

CryptographyAlgorithms Policy

StrategyDevelopment

Figure 2 General schematic of the conceptual model of the cipher algorithm design

4 Computational Intelligence and Neuroscience

(i) e relationship matrix of support indices withmain design and implementation processes of thecipher algorithms since there are 13 support indicesin the level-3 model (Figure 3) and 4 steps in thelevel-1 model (Figure 2) a 13lowast4 matrix is consti-tuted to determine the relationship between themembers of these two processes where its rows arethe elements of the support process and its columnsare the quadruple elements of the design andimplementation of cryptography algorithms eelements of this matrix are between 0 and 10 asshown in Table 2 [35] e value of each elementrepresents the effectiveness of each support index oneach design and implementation step Analysis ofthis matrix and its modeling computation providesthe possibility for the policy-makers and developersto manage the resources required for each step ofdesign and implementation of the cryptographyalgorithms and use the available hardware andsoftware resources optimally

(ii) e relationship matrix of the objectives with themain design process of the cryptography algo-rithms considering the seven objectives extracted in

Section 2 and four steps in the main process todetermine the role of each step in achieving theseven objectives a 7lowast4 matrix is constituted Sincethe number of final resources used to extract data is814 papers technical reports and documents theelements of this matrix are between 0 and 814 asgiven in Table 2 In this matrix the value of eachelement represents the number of studies docu-ments and reports indicating the relationship be-tween two components An interesting point in thismatrix is that all algorithms implemented in thestudied documents are evaluated

(iii) e relationship matrix of the implementationtechniques and the design objectives of the cryp-tography algorithms considering the 29 extractedtechniques for cryptography algorithm imple-mentation and seven main objectives of the cryp-tography algorithms a 29lowast 7 matrix is constitutedto determine how much each technique is used forcryptography implementation to achieve each ob-jective (Table 3) e elements of this matrix arebetween 0 and 814 A part of this matrix is shown inTable 4

ICT

RampD

Financial

Rule amp Regulation

Standard

Human resourceEducation

OrganizationStructure

Management

Cultural

International relation

Infrastructure

Material

equipments

Hash Function

ControllerProcess

Product

Policy

Requ

est

Plai

n Te

xt

Cipher Text

Supportive

Main Processes

Software

Symmetric Algorithm

Asymmetric Algorithm

Hardware

Figure 3 e cybernetic model of design and implementation of cipher algorithms at level 3

Computational Intelligence and Neuroscience 5

311 Design of Cryptography Algorithms Using the Cyber-netic Supermodel Considering the above discussion andpresence of numerous indices and components in thedesign and implementation of the cryptography algorithmwhich make it a complex system identifying the rela-tionship between these indices and ranking them is anecessity Accordingly in this section the Q-analysis

method is used and the output of the three matrices isanalyzed According to the level three of the proposedcybernetic model the design and implementation supportprocess of the cryptography algorithms includes 13 com-ponents On the contrary the design and implementationsteps of the cryptography algorithms also include foursteps constituting a 13 lowast 4 matrix

Table 2 e 13-component interaction matrix of the support and the main design and implementation processes of cryptographyalgorithms

Main process Level 1Cryptography algorithms Level 2

Application eoretical basis Implementation Evaluation Level 3

Support process

Software

Culture 5 6 5 2 Organizationstructure 8 7 7 7

International and public relations 5 4 4 7 Financial resources 7 6 7 6

Human resources and education 8 10 10 9 Research and development 10 8 8 10

Rules 7 4 5 10 Management 9 7 8 7

FAVA 10 5 8 7 Standard 10 3 8 10

HardwareEquipment 8 6 7 8

Infrastructure 9 5 5 7Material 7 2 5 5

Table 3 e relationship matrix of the objectives with the design process of the cryptography algorithms

Main process objectives Application eoretical basis Implementation EvaluationSecurity 203 386 516 516Simplicity 5 23 38 38Resources 41 30 81 81Flexibility 15 28 45 45Scalability 6 11 22 22Speed 72 146 247 247Reliability 2 10 11 11

Table 4 A part of the interaction matrix between the cryptography algorithm implementation techniques and the seven objectives

Level 3 Security Simplicity Using resources Flexibility Scalability Speed Reliability

Support process

Avalanche effect 32 0 0 0 0 0 0Digital signature 1 0 0 0 0 0 0

Block size 1 1 0 0 2 0 0Image sharing 0 0 0 1 0 0 0

Parallel processing 0 0 0 1 12 0 0reshold technique 3 1 0 0 0 0 0

Data mining 1 0 0 0 0 0 0Binary tree 1 0 0 0 3 0 0

Cycle 5 2 0 0 12 0 0Multistage crypto 1 0 0 0 0 0 0Hybrid method 30 3 0 2 9 0 2

Hardware 0 0 7 0 0 0 0

6 Computational Intelligence and Neuroscience

32 Calculating the Incidence Matrix First the incidencematric is obtained based on the data matrix 4ndash1 is matrixrepresents the ldquoimpact of support indices on the main designprocess of the cryptography algorithmsrdquo e data matrix iscomprised of two sets D support indices set C and qua-druple design and implementation steps (Table 5) e in-cidence matrix calculated using the data matrix for α 70is represented in Table 6 By assigning different values to theparameter α difference incidence matrices are obtainederesults of the Q-analysis using C++ coding for α 70 aregiven in Figure 4

D d1 d2 d131113864 1113865

C c1 c2 c3 c41113864 1113865(1)

321 Geometric Representation Multidimensional proper-ties of the system are defined by a simple or complex setKD(C λ) such that the entities of the set D represent thesupport indices and entities of the set C represent thequadruple design and implementation steps of the cryp-tography algorithms

In the sample with α70 7 dis is as follows

d1

d2 c1 c2 c3 c41113864 1113865

d3 c41113864 1113865

d4 c1 c31113864 1113865

d5 c1 c2 c3 c41113864 1113865

d6 c1 c2 c3 c41113864 1113865

d7 c1 c41113864 1113865

d8 c1 c2 c3 c41113864 1113865

d9 c1 c3 c41113864 1113865

d10 c1 c3 c41113864 1113865

d11 c1 c3 c41113864 1113865

d12 c1 c41113864 1113865

(2)

e simplexes of σp(di) are also

σminus1 d1( 1113857 σ3 d2( 1113857 σ0 d3( 1113857 σ1 d4( 1113857 σ3 d5( 1113857 σ0 d3( 1113857 σ1 d7( 1113857

σ3 d8( 1113857 σ2 d9( 1113857 σ2 d10( 1113857 σ2 d11( 1113857 σ1 d12( 1113857 σ1 d13( 1113857 (3)

erefore the complex dimension is 3 In other wordsthe diagnosis classes d2 (structureorganization) d5 (humanresourceseducation) d6 (research and development) andd8 (management) have the largest dimension

33 CalculatingDimensions andQ-Link Q-link is defined asthe link between a subset with smallest interface between twosubsequent dis in the chain of d1 to dn Q-link between twosubsequent dis with α70 7 is

Table 5 Sets of dis and cis support indices based on the datamatrix

d1 Culture

d2 Structureorganizationd3 Internationalpublic relationsd4 Financiald5 Human resourceseducationd6 Research and developmentd7 Rulesd8 Managementd9 FAVAd10 Standardd11 Equipmentd12 Infrastructured13 MaterialC1 ApplicationC2 eoretical basisC3 ImplementationC4 Evaluation

Table 6 e incidence matrix of the support indicesrsquo impact of thedesign steps of the cryptography algorithms with α 70

C1 C2 C3 C4

d1 0 0 0 0d2 1 1 1 1d3 0 0 0 1d4 1 0 1 0d5 1 1 1 1d6 1 1 1 1d7 1 0 0 1d8 1 1 1 1d9 1 0 1 1d10 1 0 1 1d11 1 0 1 1d12 1 0 0 1d13 1 0 0 0

Computational Intelligence and Neuroscience 7

σminus1 d1( 1113857 σ3 d2( 1113857⟶ 1 σ3 d2( 1113857 σ0 d3( 1113857⟶ 0 σ0 d3( 1113857 σ1 d4( 1113857⟶ minus 1

σ1 d4( 1113857 σ3 d5( 1113857⟶ 1 σ3 d5( 1113857 σ3 d6( 1113857⟶ 3 σ3 d6( 1113857 σ1 d7( 1113857⟶ 1

σ1 d7( 1113857 σ3 d8( 1113857⟶ 1 σ3 d8( 1113857 σ2 d9( 1113857⟶ 2 σ2 d9( 1113857 σ2 d10( 1113857⟶ 2

σ2 d10( 1113857 σ2 d11( 1113857⟶ 2 σ3 d11( 1113857 σ1 d12( 1113857⟶ 1 σ1 d12( 1113857 σ0 d13( 1113857⟶ 0

(4)

e maximum link dimension is 3 indicating the re-lationship between the diagnosis classes

331 Calculating the Structure Vectors As mentioned inthe definitions the vector Qq is a simplification basiscreated to eliminate the additional impacts in the equiv-alent simplex sets e maximum complex dimension withα70 7 is 3 erefore the first structure vector based onthe output is

Dimension 3 2 1 0 (5)

e second structure vector P is

dimensions 3 2 1 0

P Pdim3 Pdim2 Pdim1 Pdim0( 1113857

P 4 7 10 12( 1113857

(6)

where Pq is the number simplexes greater than or equal to qin the set K in which P is the number of simplex linkrepetitions (support indices) in the quadruple design andimplementation steps of the cryptography algorithms Basedon the values of these two vectors it is seen that the rela-tionship between the support indices and the quadrupledesign and implementation steps of the cryptography al-gorithms is high is issue indicates the role of supportcomponents in the design and implementation of the

cryptography algorithms which should be consideredseriously

332 Calculating the Obstruction or Flexibility VectorQlowastK represents the number of structural obstructions forsimplex interactions in dimension k

Qlowast

Q minus I⟶ Qlowast

1 1 1 11113858 1113859 minus 1 1 1 11113858 1113859⟶ Qlowast

0 0 0 01113858 1113859

(7)

As can be seen there is no obstruction in any of thecommunication levels indicating that there is a significantrelationship between the support components at eachequivalence class

333 Calculating Irregularity e value of (eccprime(σ)) iscalculated using the Chinese method e results for α70

7 are shown in Table 7 e calculated irregularity valueshows that the indices d2 (structureorganization) d5 (hu-man resourceseducation) d6 (research and development)and d8 (management) affect other indices

334 Calculating Complexity Also the results ofQ-analysiscan be used to describe structure complexity According toequations (4)ndash(9) and for α = 7 the complexity measure is

-1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1

3 0 1 3 3 1 3 2 2 2 1

0 -1 0 0 0 0 0 0 0 0

1 1 1 0 1 1 1 1 0

3 3 1 3 2 2 2 1

3 1 3 2 2 2 1

3 2 2 2 1

2 2 2 1

2 2 1

2 1

1

1 1 1 1 1 1

-1

-1

0

0

0

0

0

0

0

0

0

0

0q Q3210

11

11

SETSX2 X5 X6 X8X2 X5 X6 X8 X9 X10 X11X2 X4 X5 X6 X7 X8 X9 X10 X11 X12X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13

Figure 4 Implementation results of the model for support components of the cryptography algorithmsrsquo design with α 70

8 Computational Intelligence and Neuroscience

Q 1 1 1 1( 1113857

ψ(K) 2(1 + 2 + 3 + 4)

(4lowast 5)1113890 1113891 1

(8)

Since in the above model there is no obstruction amongthe component of the equivalent class it was expected thatthe complexity of the support components is not high andthe obtained complexity index of 1 verifies this expectatione system complexity for different alpha-cuts is shown inFigure 5

335 Ranking the Support Components of the Design andImplementation of the Cryptography Algorithms e resultsof using A-analysis are shown in Table 7 e connectionstrength of the factors in one group is specified with alpha-cut erefore the support components are grouped in 5levels Each level describes the priority and importance of thegroup in developing the cryptography algorithms In Fig-ure 6 the ranking pyramid of the support components usingQ-analysis is shown To allocate proper resources thecomponents existing in higher levels of the pyramid (Fig-ure 6) are of higher priority

336 Validation of the Results In this section the resultsof the cybernetic model and Q-analysis for supportcomponentsrsquo ranking are compared with the results re-ported in the global cybersecurity index (GCI) in 20152017 and 2018 presented by ITU [36ndash39] e GCI reportsare focused on five indices including ldquolegal cases orga-nization necessities technical issues capacity buildingand cooperationrdquo and the subindices include legaltechnical organization capacity building and coopera-tion According to the presented indices and subindices itis clear that the ldquorulesrdquo ldquostandardrdquo ldquoresearch and de-velopmentrdquo ldquoeducationrdquo and ldquomanagementrdquo are ofhigher priority in security establishment Although ourresearch is more skilled and detailed compared to the GCIreports but the results verify our findings Also Table 8shows the ranking of the support components obtainedusing Q-analysis

337 Executing the Model and Analyzing the Results of theObjectivesrsquo Impact on Design Steps of the Cryptography Al-gorithmsrsquo Matrix In this section the role of the sevencomponents on the quadruple design and implementationsteps of the cryptography algorithm is analyzed with Q-analysis Using the Q-analysis method and the 7lowast4 matrixobtained from the relationship of the cryptography algo-rithmsrsquo design objectives on their quadruple steps theirindices are ranked

338 Calculating the Incidence Matrix and Executing theModel First the incidence matrix is obtained for the datamatrix shown in Figure 2 It can be seen in Table 9 that eachelement of the two sets represents which indice e inci-dence matrix calculated from the data matrix for α5 40 is

given in Table 10 e results of implementing the model forα5 40 are given in Figure 7

4 Geometric Representation

In the sample with α5 40 dis are

d1 c1 c2 c3 c41113864 1113865

d2

d3 c1 c3 c41113864 1113865

d4 c3 c41113864 1113865

d5

d6 c1 c2 c3 c41113864 1113865

d7

(9)

e simplexes of σp(di) are also

σ3 d1( 1113857σminus1 d2( 1113857σ2 d3( 1113857σ1 d4( 1113857σminus1 d5( 1113857σ3 d6( 1113857σminus1 d7( 1113857

(10)

erefore the complex dimension is 3 In other wordsthe discriminant classes d1 (security) and d6 (speed) have thelargest dimension

41 Calculating the Structure Vectors As mentioned thevector Qq is a simplification basis to eliminate the additionaleffects in the set of equivalent simplexes e maximumcomplex dimension for α5 40 is 3 erefore the firststructure vector based on the output is

Dimension 3 2 1 1

Q 1 1 1 1( 1113857

Dimension 3 2 1 0

P 2 3 4 7( 1113857

(11)

e second structure vector P is

P 2 3 4 7( 1113857 (12)

42 Calculating the Obstruction Vector or InflexibilityQlowastK represents the number of structural obstructions forsimplex interactions in dimension k which is as follows forα5 40

1 1 1 1 1332

005

115

225

com

plex

ity

alpha cutα=5 α=6 α=7 α=8 α=9 α=10

Figure 5 System complexity of the support indices of design andimplementation of the cryptography algorithms for different alpha-cuts

Computational Intelligence and Neuroscience 9

Qlowast

Q minus I⟶ Qlowast

2 1 1 31113858 1113859 minus 1 1 1 11113858 1113859⟶ Qlowast

0 0 0 01113858 1113859

(13)

According to the obtained values it can be concludedthat there are no structural obstructions among the main

design indices of the cryptography algorithms at ismultiple objectives are considered simultaneously by thecryptography algorithm designers

Table 7 e irregularity of the data matrix parameters for α= 7

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d2 d5 d6 d8 qi 3 2 1 364 3 061d9 d10d11 qi 2 1 064 3 011d4 d7 d12 qi 1 014 3 002

human resourceRampD

rules FAVA standard

management infrastructure

organizationstructure equipmentinternational and public relations finanacial

resources and material

Figure 6 e ranking pyramid of the support components using Q-analysis

Table 8 Ranking of the support components using Q-analysis

Relationship of the support components with the design and implementation of the cryptography algorithms (q 0)No relationship α 0 complete relationship α 100

Human resource RampD rules FAVA and standard α100 10Management and infrastructure α90 9Organizationstructure and equipment α80 8International and public relations financial resources and material α70 7Culture α60 6All components α60 5

Table 9 Set of dis and cis the objectivesrsquo indices

eoretical basis C2 Reliability d7 Scalability d5 Resources d3 Security d1Implementation C3 Application C1 Speed d6 Flexibility d4 Simplicity d2Evaluation C4

Table 10 e incidence matrix of the design objectives of thecryptography algorithms with α= 40

c1 c2 c3 c4

d1 1 1 1 1d2 0 0 0 0d3 1 0 1 1d4 0 0 1 1d5 0 0 0 0d6 1 1 1 1d7 0 0 0 0

-1 -1 -1 -1 -1

2 1 -1 2

1 -1 1

-1 -1

3

3 -1 2 1 -1 3 -1

-1

-1

-1

-1

-1

-1q Q3210

11

11

SETSX1X6X1X3X6X1X3X4X6X1X3X4X6

Figure 7 e results of executing the model for design objectivesrsquoindices with α 40

10 Computational Intelligence and Neuroscience

43 Calculating Irregularity e results of applying theChinese method (ecclsquo(σ)) for calculating irregularity forα5 40 are shown in Table 11 e calculated irregu-larity value shows that indices d1 (security) and d6(speed) have received more attention compared to otherindices

44 Calculating Complexity Structural complexity for asample with α5 40 is

dimensions 3 2 1 0

Q Qdim3 Qdim2 Qdim1 Qdim0( 1113857

Q 1 1 1 1( 1113857

ψ(K) 2(1 + 2 + 3 + 4)

(4lowast 5)1113890 1113891 1

(14)

Since in the above model there was no obstructionamong the components of the equivalent classes in any of thecommunication levels it was expected that there is not ahigh complexity among objectives of the cryptography al-gorithm design and the complexity index of 1 verified thisexpectation e system complexity for different alpha-cutsis shown in Figure 8 As can be seen the system complexity is1 for all significant values of alpha

45 Prioritizing the Parameters of the Main Objectives of theCryptography Algorithm Design Using Q-AnalysisAccording to the analysis results the parameters can beclassified into 5 levelsese levels are shown in Table 12 andFigure 9 Each level indicates a priority and importance ofthe group in the development of cryptography algorithmsTo allocate proper resources the components at the higherlevels of the pyramid are of higher priority As can be seenthree objectives of security speed and optimal usage ofresources have the highest priority for the design of cryp-tography algorithms

46 Executing the Model and Analyzing the Results for theCryptography Algorithm Implementation Techniques einteraction matrix between the 29 extracted techniques andthe sevenmain objectives is shown in Figure 10e purposeof this section is to rank the cryptography algorithmimplementation techniques to achieve the goals of interest

47Calculating the IncidenceMatrixandExecuting theModele data matrix A is comprised of two sets e set Drepresents the employed techniques and the set C representsthe seven main design objectives (Table 13)

D d1 d2 d291113864 1113865

C c1 c2 c71113864 1113865(15)

Some parts of the incidence matrix calculated from thedata matrix A for α3 are shown in Table 14 and the resultsof theQ-analysis model for α3 24 are shown in Figure 11

48 Geometric Representation In the sample with α3 24the dis are

d1 c11113864 1113865

d2

d3

d4

d5

d6

d7

d8

d9

d10

d11 c11113864 1113865

d12

d13 c51113864 1113865

d14

d15 c1 c4 c51113864 1113865

d16

d17

d18

d19 c11113864 1113865

d20

d21

d22 c31113864 1113865

d23

d24

d25

d26

d27 c11113864 1113865

d28

d29

(16)

e simplexes of σp(di) are

Computational Intelligence and Neuroscience 11

σ0 d1( 1113857

σminus1 d2( 1113857

σminus1 d3( 1113857

σminus1 d4( 1113857

σminus1 d5( 1113857

σminus1 d6( 1113857

σminus1 d7( 1113857

σminus1 d8( 1113857

σminus1 d9( 1113857

σminus1 d10( 1113857

σ0 d11( 1113857

σminus1 d12( 1113857

σ0 d13( 1113857

σminus1 d14( 1113857

σ2 d15( 1113857

σminus1 d16( 1113857

σminus1 d17( 1113857

σminus1 d18( 1113857

σ0 d19( 1113857

σminus1 d20( 1113857

σminus1 d21( 1113857

σ0 d22( 1113857

σminus1 d23( 1113857

σminus1 d24( 1113857

σminus1 d25( 1113857

σminus1 d26( 1113857

σ0 d27( 1113857

σminus1 d28( 1113857

σminus1 d29( 1113857

(17)

erefore the complex dimension is 2 In other wordsthe discriminant class d15 (basic sciences) has the largestdimension

49 Calculating the Dimensions and the Q-Link In the alphadefined for cryptography algorithm implementation tech-niques to achieve the defined goals the obtained q-linkshows that these techniques are relatively independent al-though there is a weak relationship between some tech-niques Q-link in the samples with α3 24 between eachtwo subsequent di is as follows

σ0 d1( 1113857 σminus1 d2( 1113857⟶ minus 1

σminus1 d2( 1113857 σminus1 d3( 1113857⟶ minus 1

σ0 d3( 1113857 σminus1 d4( 1113857⟶ minus 1

σminus1 d4( 1113857 σminus1 d5( 1113857⟶ minus 1

σminus1 d5( 1113857 σminus1 d6( 1113857⟶ minus 1

σminus1 d6( 1113857 σminus1 d7( 1113857⟶ minus 1

σminus1 d7( 1113857 σminus1 d8( 1113857⟶ minus 1

σminus1 d8( 1113857 σminus1 d9( 1113857⟶ minus 1

σminus1 d9( 1113857 σminus1 d10( 1113857⟶ minus 1

σminus1 d10( 1113857 σ0 d11( 1113857⟶ minus 1

σ0 d11( 1113857 σminus1 d12( 1113857⟶ minus 1

σminus1 d12( 1113857 σ0 d13( 1113857⟶ minus 1

σ0 d13( 1113857 σminus1 d14( 1113857⟶ minus 1

σminus1 d14( 1113857 σ2 d15( 1113857⟶ minus 1

σ2 d15( 1113857 σminus1 d16( 1113857⟶ minus 1

σminus1 d16( 1113857 σminus1 d17( 1113857⟶ minus 1

σminus1 d17( 1113857 σminus1 d18( 1113857⟶ minus 1

σminus1 d18( 1113857 σ0 d19( 1113857⟶ minus 1

σ0 d19( 1113857 σminus1 d20( 1113857⟶ minus 1

σminus1 d20( 1113857 σminus1 d21( 1113857⟶ minus 1

σminus1 d21( 1113857 σ0 d22( 1113857⟶ minus 1

σ0 d22( 1113857 σminus1 d23( 1113857⟶ minus 1

σminus1 d23( 1113857 σminus1 d24( 1113857⟶ minus 1

σminus1 d24( 1113857 σminus1 d25( 1113857⟶ minus 1

σminus1 d25( 1113857 σminus1 d26( 1113857⟶ minus 1

σminus1 d26( 1113857 σ0 d27( 1113857⟶ minus 1

σ0 d27( 1113857 σminus1 d28( 1113857⟶ minus 1

σminus1 d28( 1113857 σminus1 d29( 1113857⟶ minus 1

(18)

Table 11 Irregularity of the cryptography objectivesrsquo parameters in the data matrix A for α5

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d1 and d6 qi 3 2 and 1 433 3 072d3 qi 2 and 1 133 3 026d4 qi 1 033 3 006

11117

1 1 1

09095

1105

11115

12

com

plex

ity

alpha cutα=2 α=3 α=5 α=15 α=25

Figure 8 System complexity of the cryptography algorithm designobjectives for different alpha-cuts

12 Computational Intelligence and Neuroscience

410 Calculating the Structure Vectors erefore the firststructure vector based on the software output is

Q 1 1 2( 1113857 (19)

e second structure vector P is

P 1 1 29( 1113857 (20)

Since the large values of P of higher dimensions dem-onstrate more links the second structure vector calculatedfor the alpha of interest shows that the relationship betweenthe cryptography algorithm implementation techniques isminimum

411 Calculating the Obstruction or Inflexibility Vectore obstruction vector (Qlowast) for a sample with α3 24 is

Table 12 Ranking the objectivesrsquo parameters considering different cuts in Q-analysis

e equivalence class for the parameters with minimum relation level (q 0) αSecurity α50 244Speed α30 244Resources α15 122Flexibility α5 40Simplicity-scalability α2 16(All parameters)

securityspeed

optimal usage of resources

flexibility

simplicity-scalability

Figure 9 Prioritizing the design and implementation objectives of the cryptography algorithms using Q-analysis

securityspeed

optimal usage of resources

flexibility

simplicity-scalability-reliability

Figure 10 Prioritizing the design and implementation objectives of the cryptography algorithms using Q-analysis

Table 13 Set of dis and cis the techniques employed for implementing cryptography algorithms

Flexibility C4 Fuzzy logic d25 Storage space d17 Cycle d9 Avalanche effect d1Scalability C5 Software d26 Clustering d18 Multiple step d10 Digital signature d2Speed C6 Steganography d27 Key d19 Hybrid method d11 Block size d3Reliability C7 Music harmony d28 Graph d20 Hardware d12 Image sharing d4

Artificial intelligence d29 Characteristic oriented d21 Hardware d13 Parallel processing d5Security C1 Energy consumption d22 Occupation area d14 reshold technique d6Simplicity C2 Bandwidth consumption d23 Basic science d15 Data mining d7Resources C3 Memory consumption d24 Compression d16 Binary tree d8

Table 14 e incidence matrix of the cryptography algorithmimplementation techniques with α 24

C1 C2 C3 C4 C5 C6 C7

d1 1 0 0 0 0 0 0d2 0 0 0 0 0 0 0d3 0 0 0 0 0 0 0d4 0 0 0 0 0 0 0d5 0 0 0 0 0 0 0d6 0 0 0 0 0 0 0d7 0 0 0 0 0 0 0d8 0 0 0 0 0 0 0d9 0 0 0 0 0 0 0d10 0 0 0 0 0 0 0d11 1 0 0 0 0 0 0d12 0 0 0 0 0 0 0

Computational Intelligence and Neuroscience 13

Qlowast

Q minus I⟶ Qlowast

1 1 21113858 1113859 minus 1 1 11113858 1113859⟶ Qlowast

0 0 11113858 1113859

(21)

e value of QlowastK represents the number of structurallimitations or obstructions for the cryptography techniquesrsquointeraction at dimension k As can be seen from the cal-culated obstruction vector there is a significant relationshipat some levels

412 Calculating Irregularity Irregularity is the integrationdegree of a cryptography method in the total complex Mea-suring irregularity (eccrsquo) for α3 24 is shown in Table 15 Ascan be seen irregularity for the discriminant class d15 indicatesthat this technique is isolated from other techniques

413 Calculating Complexity Structure complexity for asample with α3 24 is

Q 1 1 2( 1113857

ψ(K) 2(2 + 2 + 3)

(3lowast 4)1113890 1113891 117

(22)

e system complexity for different values of alpha isshown in Figure 12 As can be seen the total complexity ofthe system for large values alpha tends to stability

414 Ranking the Cryptography Algorithm ImplementationTechniques ese techniques can be classified into 5 levelsusing the analysis of the obtained results ese levels can beseen in Table 16 and Figure 13 e link power of the factorsin one group is specified with alpha Each level describespriority and importance of the group in the development ofcryptography algorithms Accordingly six methods of basicscience key management using hardware methods usingsteganography Avalanche effect and using hybrid method

q Q21o

112

SETSX15X15X1X11X15X19X27X13 amp X22

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-10-1-1-10-1-1-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-10-1-1-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

2-10-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1

-1-1-1-1-1-1

-1-1-1-1-1

-1-1-1-1

-1-1-1

-1-10

Figure 11 Results of executing the model for cryptography algorithm implementation techniques with α3 24

14 Computational Intelligence and Neuroscience

are the most important cryptography algorithm imple-mentation techniques to achieve the main objectives

5 Conclusion

In this study the design and implementation model of thecryptography algorithms is designed in three levels and itseffective components are extracted To organize the com-ponents in addition to elimination of the additional com-ponents through the measure reduction algorithm a properclassification is applied to examine the mutual effects Afterclassification three 13lowast 4 7lowast 4 and 29lowast 7 matrices areconstituted and the model is implemented on these threematrices To implement the designed model Q-analysis isused Accordingly for support indices five indices of highpriority include human resources research and develop-ment management organizationstructure and equipmentFor the algorithm design objectives index five high priorityindices include security speed optimal usage of resources

and simplicity For the indices related to implementationtechniques of the cryptography algorithms the most appliedtechniques for achieving the determined objectives includeusing basic science hardware and software methods usingkeymanagement hybrid method steganography Avalancheeffect and using artificial intelligence In the future work wewill consider the following cases

A plan should be formulated according to the prioritiesAccording to the outputs of this study and the presented pri-orities the following topics can be investigated in future studies

(1) Presenting a comprehensive model for generatingvarious cryptography algorithms based on the pri-orities of interest

(2) Examining and formulating a model about the roleof basic science for design and implementation ofcryptography algorithms considering the significantrole of ldquobasic sciencerdquo in implementation of thecryptography algorithms and about the role ofschools and universities

(3) Presenting a model for design and implementationof cryptography algorithms in the IoT infrastructurewith optimal resource usage

Data Availability

e data used to support the findings of the study areavailable from the corresponding author upon request

Conflicts of Interest

e authors declare that there are no conflicts of interestregarding the publication of this article

Table 15 Irregularity of the cryptography algorithm implementation techniques for α3 24

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d15 qi 2 1 3 2 1- qi 1 1 2 033

11117

1 1 1

091

1112

com

plex

ity

alpha-cutα=2 α=3 α=5 α=15 α=25

Figure 12 System complexity of cryptography algorithm implementation techniques for different alpha-cuts

Table 16 Ranking the implementation techniques of the cryptography algorithms considering various alpha-cuts using Q-analysis

Equivalence classes for the parameters with minimum relationship level (q 0) αBasic sciences α20 163Key α10 81Hardware and steganography α5 40Avalanche effect and hybrid method α3 24Artificial intelligence software occupation area energy consumption memory consumption α2 16

basic sciences

key

Avalanche effect hybrid method

hardware steganography

artificial intelligence soware occupationarea energy consumption memory consumption

Figure 13 Prioritizing the implementation techniques of thecryptography algorithms using Q-analysis

Computational Intelligence and Neuroscience 15

Acknowledgments

is work was partially supported by Iran National ScienceFoundation (INSF) under contract No 9653979

References

[1] D Nilesh and M Nagle ldquoe new cryptography algorithmwith high throughputrdquo in Proceedings of the 2014 Interna-tional Conference on Computer Communication and Infor-matics pp 3ndash7 Coimbatore India January 2014

[2] S Ravi P Kocher R Lee G McGraw and A RaghunathanldquoSecurity as a new dimension in embedded system designrdquo inProceedings of the 41st annual conference on Design auto-mation-DAC lsquo04 pp 753ndash760 San Diego CA USA March2004

[3] P Kuppuswamy and S Q Y A Khalidi ldquoHybrid encryptiondecryption technique using new public key and symmetric keyalgorithmrdquo International Journal of Information and Com-puter Security vol 6 no 4 pp 372ndash382 2014

[4] S Bhat and V Kapoor ldquoSecure and efficient data privacyauthentication and integrity schemes using hybrid cryptog-raphyrdquo International Conference on Advanced ComputingNetworking and Informatics vol 870 pp 279ndash285 2019

[5] D Ott and C Peikert ldquoAnd other workshop participantsldquoidentifying research challenges in post quantum cryptog-raphy migration and cryptographic agilityrdquo 2019 httparxivorgabs190907353

[6] P Manickam K Shankar E Perumal M Ilayaraja andK Sathesh Kumar ldquoSecure data transmission through reliablevehicles in VANET using optimal lightweight cryptographyrdquoin Advanced Sciences and Technologies for Security Applica-tions pp 193ndash204 Springer New York NY USA 2019

[7] S Prakash and A Rajput ldquoHybrid cryptography for securedata communication in wireless sensor networksrdquo Advancesin Intelligent Systems and Computing vol 696 pp 589ndash5992018

[8] A A Soofi I Riaz and U Rasheed ldquoAn enhanced vigenerecipher for data securityrdquo International Journal of Scientific ampTechnology Research vol 4 no 8 pp 141ndash145 2015

[9] N Shamsuddin M Syafiqah and S Ali Pitchay ldquoImple-menting location-based cryptography on mobile applicationdesign to secure data in cloud storagerdquo Journal of PhysicsConference Series IOP Publishing vol 1551 no 1 2020

[10] L Jiao H Yonglin and D Feng ldquoStream cipher designs areviewrdquo Science China Information Sciences vol 63 no 3pp 1ndash25 2020

[11] NIST ldquoNIST cryptographic standards and guidelines devel-opment processrdquo Inside NIST vol 27 2016

[12] G Alagic J Alperin-Sheriff D Apon et al Status Report onthe First Round of the NIST Post-Quantum CryptographyStandardization Process National Institute of Standards andTechnology Gaithersburg MD USA pp 1ndash27 2019

[13] I Damaj and S Kasbah ldquoAn analysis framework for hardwareand software implementations with applications from cryp-tographyrdquo Computers amp Electrical Engineering vol 69pp 572ndash584 2018

[14] M S Croock Z A Hassan and S D Khudhur ldquoAdaptive keygeneration algorithm based on software engineering meth-odologyrdquo International Journal of Electrical and ComputerEngineering vol 11 no 1 p 589 2021

[15] K Sha T A Yang W Wei and S Davari ldquoA survey of edgecomputing-based designs for iot securityrdquo Digital Commu-nications and Networks vol 6 no 2 pp 195ndash202 2020

[16] M Yahuza M Y I B Idris A W B A Wahab et alldquoSystematic review on security and privacy requirements inedge computing state of the art and future research oppor-tunitiesrdquo IEEE Access vol 8 pp 76541ndash76567 2020

[17] A Bhowmik J Dey and S Karforma A Novel Double TierCryptographic System (nDTCS) to Reinforce Patientsrsquo Privacyin Contemporary COVID-19 Telemedicine Springer NewYork NY USA 2021

[18] N Gupta User Integrity Protection Security Model forEnhanced Telemedicine for Healthcare Networks 2020httpsosfio69jdq

[19] J Dey A Bhowmik A Sarkar S Karforma andB Chowdhury Cryptographic Engineering on COVID-19Telemedicine An Intelligent Transmission through RecurrentRelation Based Session Key Springer New York NY USA2021

[20] V Hosseinian H Ayatollahi H Haghani and E MehraeenldquoRequirements of information security in a telemedicinenetwork review of IT managersrsquo opinionrdquo Journal of Para-medical Sciences amp Rehabilitation vol 4 no 2 pp 31ndash402015

[21] D Karaoglan Altop A Levi and V Tuzcu ldquoDeriving cryp-tographic keys from physiological signalsrdquo Pervasive andMobile Computing vol 39 pp 65ndash79 2017

[22] O Uzunkol and M S Kiraz ldquoStill wrong use of pairings incryptographyrdquo Applied Mathematics and Computationvol 333 pp 467ndash479 2018

[23] S Picek D Jakobovic J F Miller L Batina and M CupicldquoCryptographic Boolean functions one output many designcriteriardquo Applied Soft Computing vol 40 pp 635ndash653 2016

[24] R Saha and G Geetha ldquoSymmetric random function gen-erator (SRFG) a novel cryptographic primitive for designingfast and robust algorithmsrdquo Chaos Solitons amp Fractalsvol 104 pp 371ndash377 2017

[25] A Achuthshankar and A Achuthshankar ldquoA novel sym-metric cryptography algorithm for fast and secure encryp-tionrdquo in Proceedings of the 2015 IEEE 9th InternationalConference on Intelligent Systems and Control (ISCO) pp 4ndash9Coimbatore India January 2015

[26] D Ott and C Peikert ldquoIdentifying research challenges in postquantum cryptography migration and cryptographic agilityrdquo2019 httparxivorgabs190907353

[27] W Liu B Ying H Yang and H Wang ldquoAccurate modelingfor predicting cryptography overheads on wireless sensornodesrdquovol 2 pp 997ndash1001 in Proceedings of the 2019 11thInternational Conference on Advanced CommunicationTechnology vol 2 pp 997ndash1001 IEEE Bangalore IndiaFebruary 2019

[28] S Feizi A Ahmadi and A Nemati ldquoA hardware imple-mentation of simon cryptography algorithmrdquo in Proceedingsof the 2014 4th International Conference on Computer andKnowledge Engineering (ICCKE) pp 245ndash250 IEEEMashhad Iran 2014

[29] E ambiraja G Ramesh and R Umarani ldquoA survey onvarious most common encryption techniquesrdquo 2012 httpwwwijarcssecom

[30] A Gupta and N K Walia ldquoCryptography Algorithms areviewrdquo International Journal of Engineering Research andDevelopment vol 2 no 2 pp 1667ndash1672 2014

[31] R Bhanot and R Hans ldquoA review and comparative analysis ofvarious encryption algorithmsrdquo International Journal of Se-curity and Its Applications vol 9 no 4 pp 289ndash306 2015

[32] S Chandra S Paira S S Alam and G Sanyal ldquoA com-parative survey of symmetric and asymmetric key

16 Computational Intelligence and Neuroscience

cryptographyrdquo in Proceedings of the 2014 InternationalConference on Electronics Communication and Computa-tional Engineering (ICECCE) pp 83ndash93 Hosur India No-vember 2014

[33] F Maqsood M Ahmed M Mumtaz and M Ali ldquoCryp-tography a comparative analysis for modern techniquesrdquoInternational Journal of Advanced Computer Science andApplications vol 8 no 6 p 6 2017

[34] S B Sadkhan and A O Salman ldquoA survey on lightweight-cryptography status and future challengesrdquo in Proceedings ofthe International Conference on Advances in SustainableEngineering and Applications pp 105ndash108 Wasit Iraq June2018

[35] N V oai ldquoCriteria and dimension reduction of linearmultiple criteria optimization problemsrdquo Journal of GlobalOptimization vol 52 no 3 pp 499ndash508 2012

[36] C E Shannon ldquoCommunication theory of secrecy systems1945rdquo MD Computing vol 15 no 1 pp 57ndash64 1998

[37] L J Fennelly M Beaudry and M A Perry Security in 2025[38] ITU Global Cybersecurity Index 2018 2019[39] ITU Global Cybersecurity Index (GCI) 2017 2017

Computational Intelligence and Neuroscience 17

including design and implementation objectives of cipher al-gorithms applications of cipher algorithms and imple-mentation methods of the cipher algorithms In the followingthe reduced measures are classified into the above classes and

modeling is carried out based on available measures of thesethree classes Accordingly based on the level-3 cyberneticmodel (Figure 3) the following three matrices are constitutedto design and implement the cipher algorithms

Table 1 e importance of information security needs in telemedicine technology in the field of information transmission

Data transfer Veryimportant Important No

idea Nonsignificant

Implement network protocols to ensure the transmission of information and checkits integrity 265 430 43 0

Establish a communication protocol to share information between local healthinstitutions 50 37 13 0

Encrypt important files and information 262 635 0 22Investigation of encryption mechanism by technical team of security assessor 556 328 152 0Use combinations of numbers and letters for encryption 268 25 68 0Use uppercase and lowercase letters for encryption to access remote networknetworks 50 24 174 87

Methods for controlling the integrity of application information 353 635 89 22

Hash Function

AsymmetricAlgorithms

SymmetricAlgorithms

ImplementationBasicApplication

Ciph

er T

ext

EvaluationPlai

n te

xt

Figure 1 Main process of design and implementation of cipher algorithms at the highest level

Cipher textRequest Plain text

Main Processes

Control Processes

Process

Product

Supportive Processes

Software

Hardware

CryptographyAlgorithms Policy

StrategyDevelopment

Figure 2 General schematic of the conceptual model of the cipher algorithm design

4 Computational Intelligence and Neuroscience

(i) e relationship matrix of support indices withmain design and implementation processes of thecipher algorithms since there are 13 support indicesin the level-3 model (Figure 3) and 4 steps in thelevel-1 model (Figure 2) a 13lowast4 matrix is consti-tuted to determine the relationship between themembers of these two processes where its rows arethe elements of the support process and its columnsare the quadruple elements of the design andimplementation of cryptography algorithms eelements of this matrix are between 0 and 10 asshown in Table 2 [35] e value of each elementrepresents the effectiveness of each support index oneach design and implementation step Analysis ofthis matrix and its modeling computation providesthe possibility for the policy-makers and developersto manage the resources required for each step ofdesign and implementation of the cryptographyalgorithms and use the available hardware andsoftware resources optimally

(ii) e relationship matrix of the objectives with themain design process of the cryptography algo-rithms considering the seven objectives extracted in

Section 2 and four steps in the main process todetermine the role of each step in achieving theseven objectives a 7lowast4 matrix is constituted Sincethe number of final resources used to extract data is814 papers technical reports and documents theelements of this matrix are between 0 and 814 asgiven in Table 2 In this matrix the value of eachelement represents the number of studies docu-ments and reports indicating the relationship be-tween two components An interesting point in thismatrix is that all algorithms implemented in thestudied documents are evaluated

(iii) e relationship matrix of the implementationtechniques and the design objectives of the cryp-tography algorithms considering the 29 extractedtechniques for cryptography algorithm imple-mentation and seven main objectives of the cryp-tography algorithms a 29lowast 7 matrix is constitutedto determine how much each technique is used forcryptography implementation to achieve each ob-jective (Table 3) e elements of this matrix arebetween 0 and 814 A part of this matrix is shown inTable 4

ICT

RampD

Financial

Rule amp Regulation

Standard

Human resourceEducation

OrganizationStructure

Management

Cultural

International relation

Infrastructure

Material

equipments

Hash Function

ControllerProcess

Product

Policy

Requ

est

Plai

n Te

xt

Cipher Text

Supportive

Main Processes

Software

Symmetric Algorithm

Asymmetric Algorithm

Hardware

Figure 3 e cybernetic model of design and implementation of cipher algorithms at level 3

Computational Intelligence and Neuroscience 5

311 Design of Cryptography Algorithms Using the Cyber-netic Supermodel Considering the above discussion andpresence of numerous indices and components in thedesign and implementation of the cryptography algorithmwhich make it a complex system identifying the rela-tionship between these indices and ranking them is anecessity Accordingly in this section the Q-analysis

method is used and the output of the three matrices isanalyzed According to the level three of the proposedcybernetic model the design and implementation supportprocess of the cryptography algorithms includes 13 com-ponents On the contrary the design and implementationsteps of the cryptography algorithms also include foursteps constituting a 13 lowast 4 matrix

Table 2 e 13-component interaction matrix of the support and the main design and implementation processes of cryptographyalgorithms

Main process Level 1Cryptography algorithms Level 2

Application eoretical basis Implementation Evaluation Level 3

Support process

Software

Culture 5 6 5 2 Organizationstructure 8 7 7 7

International and public relations 5 4 4 7 Financial resources 7 6 7 6

Human resources and education 8 10 10 9 Research and development 10 8 8 10

Rules 7 4 5 10 Management 9 7 8 7

FAVA 10 5 8 7 Standard 10 3 8 10

HardwareEquipment 8 6 7 8

Infrastructure 9 5 5 7Material 7 2 5 5

Table 3 e relationship matrix of the objectives with the design process of the cryptography algorithms

Main process objectives Application eoretical basis Implementation EvaluationSecurity 203 386 516 516Simplicity 5 23 38 38Resources 41 30 81 81Flexibility 15 28 45 45Scalability 6 11 22 22Speed 72 146 247 247Reliability 2 10 11 11

Table 4 A part of the interaction matrix between the cryptography algorithm implementation techniques and the seven objectives

Level 3 Security Simplicity Using resources Flexibility Scalability Speed Reliability

Support process

Avalanche effect 32 0 0 0 0 0 0Digital signature 1 0 0 0 0 0 0

Block size 1 1 0 0 2 0 0Image sharing 0 0 0 1 0 0 0

Parallel processing 0 0 0 1 12 0 0reshold technique 3 1 0 0 0 0 0

Data mining 1 0 0 0 0 0 0Binary tree 1 0 0 0 3 0 0

Cycle 5 2 0 0 12 0 0Multistage crypto 1 0 0 0 0 0 0Hybrid method 30 3 0 2 9 0 2

Hardware 0 0 7 0 0 0 0

6 Computational Intelligence and Neuroscience

32 Calculating the Incidence Matrix First the incidencematric is obtained based on the data matrix 4ndash1 is matrixrepresents the ldquoimpact of support indices on the main designprocess of the cryptography algorithmsrdquo e data matrix iscomprised of two sets D support indices set C and qua-druple design and implementation steps (Table 5) e in-cidence matrix calculated using the data matrix for α 70is represented in Table 6 By assigning different values to theparameter α difference incidence matrices are obtainederesults of the Q-analysis using C++ coding for α 70 aregiven in Figure 4

D d1 d2 d131113864 1113865

C c1 c2 c3 c41113864 1113865(1)

321 Geometric Representation Multidimensional proper-ties of the system are defined by a simple or complex setKD(C λ) such that the entities of the set D represent thesupport indices and entities of the set C represent thequadruple design and implementation steps of the cryp-tography algorithms

In the sample with α70 7 dis is as follows

d1

d2 c1 c2 c3 c41113864 1113865

d3 c41113864 1113865

d4 c1 c31113864 1113865

d5 c1 c2 c3 c41113864 1113865

d6 c1 c2 c3 c41113864 1113865

d7 c1 c41113864 1113865

d8 c1 c2 c3 c41113864 1113865

d9 c1 c3 c41113864 1113865

d10 c1 c3 c41113864 1113865

d11 c1 c3 c41113864 1113865

d12 c1 c41113864 1113865

(2)

e simplexes of σp(di) are also

σminus1 d1( 1113857 σ3 d2( 1113857 σ0 d3( 1113857 σ1 d4( 1113857 σ3 d5( 1113857 σ0 d3( 1113857 σ1 d7( 1113857

σ3 d8( 1113857 σ2 d9( 1113857 σ2 d10( 1113857 σ2 d11( 1113857 σ1 d12( 1113857 σ1 d13( 1113857 (3)

erefore the complex dimension is 3 In other wordsthe diagnosis classes d2 (structureorganization) d5 (humanresourceseducation) d6 (research and development) andd8 (management) have the largest dimension

33 CalculatingDimensions andQ-Link Q-link is defined asthe link between a subset with smallest interface between twosubsequent dis in the chain of d1 to dn Q-link between twosubsequent dis with α70 7 is

Table 5 Sets of dis and cis support indices based on the datamatrix

d1 Culture

d2 Structureorganizationd3 Internationalpublic relationsd4 Financiald5 Human resourceseducationd6 Research and developmentd7 Rulesd8 Managementd9 FAVAd10 Standardd11 Equipmentd12 Infrastructured13 MaterialC1 ApplicationC2 eoretical basisC3 ImplementationC4 Evaluation

Table 6 e incidence matrix of the support indicesrsquo impact of thedesign steps of the cryptography algorithms with α 70

C1 C2 C3 C4

d1 0 0 0 0d2 1 1 1 1d3 0 0 0 1d4 1 0 1 0d5 1 1 1 1d6 1 1 1 1d7 1 0 0 1d8 1 1 1 1d9 1 0 1 1d10 1 0 1 1d11 1 0 1 1d12 1 0 0 1d13 1 0 0 0

Computational Intelligence and Neuroscience 7

σminus1 d1( 1113857 σ3 d2( 1113857⟶ 1 σ3 d2( 1113857 σ0 d3( 1113857⟶ 0 σ0 d3( 1113857 σ1 d4( 1113857⟶ minus 1

σ1 d4( 1113857 σ3 d5( 1113857⟶ 1 σ3 d5( 1113857 σ3 d6( 1113857⟶ 3 σ3 d6( 1113857 σ1 d7( 1113857⟶ 1

σ1 d7( 1113857 σ3 d8( 1113857⟶ 1 σ3 d8( 1113857 σ2 d9( 1113857⟶ 2 σ2 d9( 1113857 σ2 d10( 1113857⟶ 2

σ2 d10( 1113857 σ2 d11( 1113857⟶ 2 σ3 d11( 1113857 σ1 d12( 1113857⟶ 1 σ1 d12( 1113857 σ0 d13( 1113857⟶ 0

(4)

e maximum link dimension is 3 indicating the re-lationship between the diagnosis classes

331 Calculating the Structure Vectors As mentioned inthe definitions the vector Qq is a simplification basiscreated to eliminate the additional impacts in the equiv-alent simplex sets e maximum complex dimension withα70 7 is 3 erefore the first structure vector based onthe output is

Dimension 3 2 1 0 (5)

e second structure vector P is

dimensions 3 2 1 0

P Pdim3 Pdim2 Pdim1 Pdim0( 1113857

P 4 7 10 12( 1113857

(6)

where Pq is the number simplexes greater than or equal to qin the set K in which P is the number of simplex linkrepetitions (support indices) in the quadruple design andimplementation steps of the cryptography algorithms Basedon the values of these two vectors it is seen that the rela-tionship between the support indices and the quadrupledesign and implementation steps of the cryptography al-gorithms is high is issue indicates the role of supportcomponents in the design and implementation of the

cryptography algorithms which should be consideredseriously

332 Calculating the Obstruction or Flexibility VectorQlowastK represents the number of structural obstructions forsimplex interactions in dimension k

Qlowast

Q minus I⟶ Qlowast

1 1 1 11113858 1113859 minus 1 1 1 11113858 1113859⟶ Qlowast

0 0 0 01113858 1113859

(7)

As can be seen there is no obstruction in any of thecommunication levels indicating that there is a significantrelationship between the support components at eachequivalence class

333 Calculating Irregularity e value of (eccprime(σ)) iscalculated using the Chinese method e results for α70

7 are shown in Table 7 e calculated irregularity valueshows that the indices d2 (structureorganization) d5 (hu-man resourceseducation) d6 (research and development)and d8 (management) affect other indices

334 Calculating Complexity Also the results ofQ-analysiscan be used to describe structure complexity According toequations (4)ndash(9) and for α = 7 the complexity measure is

-1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1

3 0 1 3 3 1 3 2 2 2 1

0 -1 0 0 0 0 0 0 0 0

1 1 1 0 1 1 1 1 0

3 3 1 3 2 2 2 1

3 1 3 2 2 2 1

3 2 2 2 1

2 2 2 1

2 2 1

2 1

1

1 1 1 1 1 1

-1

-1

0

0

0

0

0

0

0

0

0

0

0q Q3210

11

11

SETSX2 X5 X6 X8X2 X5 X6 X8 X9 X10 X11X2 X4 X5 X6 X7 X8 X9 X10 X11 X12X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13

Figure 4 Implementation results of the model for support components of the cryptography algorithmsrsquo design with α 70

8 Computational Intelligence and Neuroscience

Q 1 1 1 1( 1113857

ψ(K) 2(1 + 2 + 3 + 4)

(4lowast 5)1113890 1113891 1

(8)

Since in the above model there is no obstruction amongthe component of the equivalent class it was expected thatthe complexity of the support components is not high andthe obtained complexity index of 1 verifies this expectatione system complexity for different alpha-cuts is shown inFigure 5

335 Ranking the Support Components of the Design andImplementation of the Cryptography Algorithms e resultsof using A-analysis are shown in Table 7 e connectionstrength of the factors in one group is specified with alpha-cut erefore the support components are grouped in 5levels Each level describes the priority and importance of thegroup in developing the cryptography algorithms In Fig-ure 6 the ranking pyramid of the support components usingQ-analysis is shown To allocate proper resources thecomponents existing in higher levels of the pyramid (Fig-ure 6) are of higher priority

336 Validation of the Results In this section the resultsof the cybernetic model and Q-analysis for supportcomponentsrsquo ranking are compared with the results re-ported in the global cybersecurity index (GCI) in 20152017 and 2018 presented by ITU [36ndash39] e GCI reportsare focused on five indices including ldquolegal cases orga-nization necessities technical issues capacity buildingand cooperationrdquo and the subindices include legaltechnical organization capacity building and coopera-tion According to the presented indices and subindices itis clear that the ldquorulesrdquo ldquostandardrdquo ldquoresearch and de-velopmentrdquo ldquoeducationrdquo and ldquomanagementrdquo are ofhigher priority in security establishment Although ourresearch is more skilled and detailed compared to the GCIreports but the results verify our findings Also Table 8shows the ranking of the support components obtainedusing Q-analysis

337 Executing the Model and Analyzing the Results of theObjectivesrsquo Impact on Design Steps of the Cryptography Al-gorithmsrsquo Matrix In this section the role of the sevencomponents on the quadruple design and implementationsteps of the cryptography algorithm is analyzed with Q-analysis Using the Q-analysis method and the 7lowast4 matrixobtained from the relationship of the cryptography algo-rithmsrsquo design objectives on their quadruple steps theirindices are ranked

338 Calculating the Incidence Matrix and Executing theModel First the incidence matrix is obtained for the datamatrix shown in Figure 2 It can be seen in Table 9 that eachelement of the two sets represents which indice e inci-dence matrix calculated from the data matrix for α5 40 is

given in Table 10 e results of implementing the model forα5 40 are given in Figure 7

4 Geometric Representation

In the sample with α5 40 dis are

d1 c1 c2 c3 c41113864 1113865

d2

d3 c1 c3 c41113864 1113865

d4 c3 c41113864 1113865

d5

d6 c1 c2 c3 c41113864 1113865

d7

(9)

e simplexes of σp(di) are also

σ3 d1( 1113857σminus1 d2( 1113857σ2 d3( 1113857σ1 d4( 1113857σminus1 d5( 1113857σ3 d6( 1113857σminus1 d7( 1113857

(10)

erefore the complex dimension is 3 In other wordsthe discriminant classes d1 (security) and d6 (speed) have thelargest dimension

41 Calculating the Structure Vectors As mentioned thevector Qq is a simplification basis to eliminate the additionaleffects in the set of equivalent simplexes e maximumcomplex dimension for α5 40 is 3 erefore the firststructure vector based on the output is

Dimension 3 2 1 1

Q 1 1 1 1( 1113857

Dimension 3 2 1 0

P 2 3 4 7( 1113857

(11)

e second structure vector P is

P 2 3 4 7( 1113857 (12)

42 Calculating the Obstruction Vector or InflexibilityQlowastK represents the number of structural obstructions forsimplex interactions in dimension k which is as follows forα5 40

1 1 1 1 1332

005

115

225

com

plex

ity

alpha cutα=5 α=6 α=7 α=8 α=9 α=10

Figure 5 System complexity of the support indices of design andimplementation of the cryptography algorithms for different alpha-cuts

Computational Intelligence and Neuroscience 9

Qlowast

Q minus I⟶ Qlowast

2 1 1 31113858 1113859 minus 1 1 1 11113858 1113859⟶ Qlowast

0 0 0 01113858 1113859

(13)

According to the obtained values it can be concludedthat there are no structural obstructions among the main

design indices of the cryptography algorithms at ismultiple objectives are considered simultaneously by thecryptography algorithm designers

Table 7 e irregularity of the data matrix parameters for α= 7

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d2 d5 d6 d8 qi 3 2 1 364 3 061d9 d10d11 qi 2 1 064 3 011d4 d7 d12 qi 1 014 3 002

human resourceRampD

rules FAVA standard

management infrastructure

organizationstructure equipmentinternational and public relations finanacial

resources and material

Figure 6 e ranking pyramid of the support components using Q-analysis

Table 8 Ranking of the support components using Q-analysis

Relationship of the support components with the design and implementation of the cryptography algorithms (q 0)No relationship α 0 complete relationship α 100

Human resource RampD rules FAVA and standard α100 10Management and infrastructure α90 9Organizationstructure and equipment α80 8International and public relations financial resources and material α70 7Culture α60 6All components α60 5

Table 9 Set of dis and cis the objectivesrsquo indices

eoretical basis C2 Reliability d7 Scalability d5 Resources d3 Security d1Implementation C3 Application C1 Speed d6 Flexibility d4 Simplicity d2Evaluation C4

Table 10 e incidence matrix of the design objectives of thecryptography algorithms with α= 40

c1 c2 c3 c4

d1 1 1 1 1d2 0 0 0 0d3 1 0 1 1d4 0 0 1 1d5 0 0 0 0d6 1 1 1 1d7 0 0 0 0

-1 -1 -1 -1 -1

2 1 -1 2

1 -1 1

-1 -1

3

3 -1 2 1 -1 3 -1

-1

-1

-1

-1

-1

-1q Q3210

11

11

SETSX1X6X1X3X6X1X3X4X6X1X3X4X6

Figure 7 e results of executing the model for design objectivesrsquoindices with α 40

10 Computational Intelligence and Neuroscience

43 Calculating Irregularity e results of applying theChinese method (ecclsquo(σ)) for calculating irregularity forα5 40 are shown in Table 11 e calculated irregu-larity value shows that indices d1 (security) and d6(speed) have received more attention compared to otherindices

44 Calculating Complexity Structural complexity for asample with α5 40 is

dimensions 3 2 1 0

Q Qdim3 Qdim2 Qdim1 Qdim0( 1113857

Q 1 1 1 1( 1113857

ψ(K) 2(1 + 2 + 3 + 4)

(4lowast 5)1113890 1113891 1

(14)

Since in the above model there was no obstructionamong the components of the equivalent classes in any of thecommunication levels it was expected that there is not ahigh complexity among objectives of the cryptography al-gorithm design and the complexity index of 1 verified thisexpectation e system complexity for different alpha-cutsis shown in Figure 8 As can be seen the system complexity is1 for all significant values of alpha

45 Prioritizing the Parameters of the Main Objectives of theCryptography Algorithm Design Using Q-AnalysisAccording to the analysis results the parameters can beclassified into 5 levelsese levels are shown in Table 12 andFigure 9 Each level indicates a priority and importance ofthe group in the development of cryptography algorithmsTo allocate proper resources the components at the higherlevels of the pyramid are of higher priority As can be seenthree objectives of security speed and optimal usage ofresources have the highest priority for the design of cryp-tography algorithms

46 Executing the Model and Analyzing the Results for theCryptography Algorithm Implementation Techniques einteraction matrix between the 29 extracted techniques andthe sevenmain objectives is shown in Figure 10e purposeof this section is to rank the cryptography algorithmimplementation techniques to achieve the goals of interest

47Calculating the IncidenceMatrixandExecuting theModele data matrix A is comprised of two sets e set Drepresents the employed techniques and the set C representsthe seven main design objectives (Table 13)

D d1 d2 d291113864 1113865

C c1 c2 c71113864 1113865(15)

Some parts of the incidence matrix calculated from thedata matrix A for α3 are shown in Table 14 and the resultsof theQ-analysis model for α3 24 are shown in Figure 11

48 Geometric Representation In the sample with α3 24the dis are

d1 c11113864 1113865

d2

d3

d4

d5

d6

d7

d8

d9

d10

d11 c11113864 1113865

d12

d13 c51113864 1113865

d14

d15 c1 c4 c51113864 1113865

d16

d17

d18

d19 c11113864 1113865

d20

d21

d22 c31113864 1113865

d23

d24

d25

d26

d27 c11113864 1113865

d28

d29

(16)

e simplexes of σp(di) are

Computational Intelligence and Neuroscience 11

σ0 d1( 1113857

σminus1 d2( 1113857

σminus1 d3( 1113857

σminus1 d4( 1113857

σminus1 d5( 1113857

σminus1 d6( 1113857

σminus1 d7( 1113857

σminus1 d8( 1113857

σminus1 d9( 1113857

σminus1 d10( 1113857

σ0 d11( 1113857

σminus1 d12( 1113857

σ0 d13( 1113857

σminus1 d14( 1113857

σ2 d15( 1113857

σminus1 d16( 1113857

σminus1 d17( 1113857

σminus1 d18( 1113857

σ0 d19( 1113857

σminus1 d20( 1113857

σminus1 d21( 1113857

σ0 d22( 1113857

σminus1 d23( 1113857

σminus1 d24( 1113857

σminus1 d25( 1113857

σminus1 d26( 1113857

σ0 d27( 1113857

σminus1 d28( 1113857

σminus1 d29( 1113857

(17)

erefore the complex dimension is 2 In other wordsthe discriminant class d15 (basic sciences) has the largestdimension

49 Calculating the Dimensions and the Q-Link In the alphadefined for cryptography algorithm implementation tech-niques to achieve the defined goals the obtained q-linkshows that these techniques are relatively independent al-though there is a weak relationship between some tech-niques Q-link in the samples with α3 24 between eachtwo subsequent di is as follows

σ0 d1( 1113857 σminus1 d2( 1113857⟶ minus 1

σminus1 d2( 1113857 σminus1 d3( 1113857⟶ minus 1

σ0 d3( 1113857 σminus1 d4( 1113857⟶ minus 1

σminus1 d4( 1113857 σminus1 d5( 1113857⟶ minus 1

σminus1 d5( 1113857 σminus1 d6( 1113857⟶ minus 1

σminus1 d6( 1113857 σminus1 d7( 1113857⟶ minus 1

σminus1 d7( 1113857 σminus1 d8( 1113857⟶ minus 1

σminus1 d8( 1113857 σminus1 d9( 1113857⟶ minus 1

σminus1 d9( 1113857 σminus1 d10( 1113857⟶ minus 1

σminus1 d10( 1113857 σ0 d11( 1113857⟶ minus 1

σ0 d11( 1113857 σminus1 d12( 1113857⟶ minus 1

σminus1 d12( 1113857 σ0 d13( 1113857⟶ minus 1

σ0 d13( 1113857 σminus1 d14( 1113857⟶ minus 1

σminus1 d14( 1113857 σ2 d15( 1113857⟶ minus 1

σ2 d15( 1113857 σminus1 d16( 1113857⟶ minus 1

σminus1 d16( 1113857 σminus1 d17( 1113857⟶ minus 1

σminus1 d17( 1113857 σminus1 d18( 1113857⟶ minus 1

σminus1 d18( 1113857 σ0 d19( 1113857⟶ minus 1

σ0 d19( 1113857 σminus1 d20( 1113857⟶ minus 1

σminus1 d20( 1113857 σminus1 d21( 1113857⟶ minus 1

σminus1 d21( 1113857 σ0 d22( 1113857⟶ minus 1

σ0 d22( 1113857 σminus1 d23( 1113857⟶ minus 1

σminus1 d23( 1113857 σminus1 d24( 1113857⟶ minus 1

σminus1 d24( 1113857 σminus1 d25( 1113857⟶ minus 1

σminus1 d25( 1113857 σminus1 d26( 1113857⟶ minus 1

σminus1 d26( 1113857 σ0 d27( 1113857⟶ minus 1

σ0 d27( 1113857 σminus1 d28( 1113857⟶ minus 1

σminus1 d28( 1113857 σminus1 d29( 1113857⟶ minus 1

(18)

Table 11 Irregularity of the cryptography objectivesrsquo parameters in the data matrix A for α5

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d1 and d6 qi 3 2 and 1 433 3 072d3 qi 2 and 1 133 3 026d4 qi 1 033 3 006

11117

1 1 1

09095

1105

11115

12

com

plex

ity

alpha cutα=2 α=3 α=5 α=15 α=25

Figure 8 System complexity of the cryptography algorithm designobjectives for different alpha-cuts

12 Computational Intelligence and Neuroscience

410 Calculating the Structure Vectors erefore the firststructure vector based on the software output is

Q 1 1 2( 1113857 (19)

e second structure vector P is

P 1 1 29( 1113857 (20)

Since the large values of P of higher dimensions dem-onstrate more links the second structure vector calculatedfor the alpha of interest shows that the relationship betweenthe cryptography algorithm implementation techniques isminimum

411 Calculating the Obstruction or Inflexibility Vectore obstruction vector (Qlowast) for a sample with α3 24 is

Table 12 Ranking the objectivesrsquo parameters considering different cuts in Q-analysis

e equivalence class for the parameters with minimum relation level (q 0) αSecurity α50 244Speed α30 244Resources α15 122Flexibility α5 40Simplicity-scalability α2 16(All parameters)

securityspeed

optimal usage of resources

flexibility

simplicity-scalability

Figure 9 Prioritizing the design and implementation objectives of the cryptography algorithms using Q-analysis

securityspeed

optimal usage of resources

flexibility

simplicity-scalability-reliability

Figure 10 Prioritizing the design and implementation objectives of the cryptography algorithms using Q-analysis

Table 13 Set of dis and cis the techniques employed for implementing cryptography algorithms

Flexibility C4 Fuzzy logic d25 Storage space d17 Cycle d9 Avalanche effect d1Scalability C5 Software d26 Clustering d18 Multiple step d10 Digital signature d2Speed C6 Steganography d27 Key d19 Hybrid method d11 Block size d3Reliability C7 Music harmony d28 Graph d20 Hardware d12 Image sharing d4

Artificial intelligence d29 Characteristic oriented d21 Hardware d13 Parallel processing d5Security C1 Energy consumption d22 Occupation area d14 reshold technique d6Simplicity C2 Bandwidth consumption d23 Basic science d15 Data mining d7Resources C3 Memory consumption d24 Compression d16 Binary tree d8

Table 14 e incidence matrix of the cryptography algorithmimplementation techniques with α 24

C1 C2 C3 C4 C5 C6 C7

d1 1 0 0 0 0 0 0d2 0 0 0 0 0 0 0d3 0 0 0 0 0 0 0d4 0 0 0 0 0 0 0d5 0 0 0 0 0 0 0d6 0 0 0 0 0 0 0d7 0 0 0 0 0 0 0d8 0 0 0 0 0 0 0d9 0 0 0 0 0 0 0d10 0 0 0 0 0 0 0d11 1 0 0 0 0 0 0d12 0 0 0 0 0 0 0

Computational Intelligence and Neuroscience 13

Qlowast

Q minus I⟶ Qlowast

1 1 21113858 1113859 minus 1 1 11113858 1113859⟶ Qlowast

0 0 11113858 1113859

(21)

e value of QlowastK represents the number of structurallimitations or obstructions for the cryptography techniquesrsquointeraction at dimension k As can be seen from the cal-culated obstruction vector there is a significant relationshipat some levels

412 Calculating Irregularity Irregularity is the integrationdegree of a cryptography method in the total complex Mea-suring irregularity (eccrsquo) for α3 24 is shown in Table 15 Ascan be seen irregularity for the discriminant class d15 indicatesthat this technique is isolated from other techniques

413 Calculating Complexity Structure complexity for asample with α3 24 is

Q 1 1 2( 1113857

ψ(K) 2(2 + 2 + 3)

(3lowast 4)1113890 1113891 117

(22)

e system complexity for different values of alpha isshown in Figure 12 As can be seen the total complexity ofthe system for large values alpha tends to stability

414 Ranking the Cryptography Algorithm ImplementationTechniques ese techniques can be classified into 5 levelsusing the analysis of the obtained results ese levels can beseen in Table 16 and Figure 13 e link power of the factorsin one group is specified with alpha Each level describespriority and importance of the group in the development ofcryptography algorithms Accordingly six methods of basicscience key management using hardware methods usingsteganography Avalanche effect and using hybrid method

q Q21o

112

SETSX15X15X1X11X15X19X27X13 amp X22

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-10-1-1-10-1-1-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-10-1-1-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

2-10-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1

-1-1-1-1-1-1

-1-1-1-1-1

-1-1-1-1

-1-1-1

-1-10

Figure 11 Results of executing the model for cryptography algorithm implementation techniques with α3 24

14 Computational Intelligence and Neuroscience

are the most important cryptography algorithm imple-mentation techniques to achieve the main objectives

5 Conclusion

In this study the design and implementation model of thecryptography algorithms is designed in three levels and itseffective components are extracted To organize the com-ponents in addition to elimination of the additional com-ponents through the measure reduction algorithm a properclassification is applied to examine the mutual effects Afterclassification three 13lowast 4 7lowast 4 and 29lowast 7 matrices areconstituted and the model is implemented on these threematrices To implement the designed model Q-analysis isused Accordingly for support indices five indices of highpriority include human resources research and develop-ment management organizationstructure and equipmentFor the algorithm design objectives index five high priorityindices include security speed optimal usage of resources

and simplicity For the indices related to implementationtechniques of the cryptography algorithms the most appliedtechniques for achieving the determined objectives includeusing basic science hardware and software methods usingkeymanagement hybrid method steganography Avalancheeffect and using artificial intelligence In the future work wewill consider the following cases

A plan should be formulated according to the prioritiesAccording to the outputs of this study and the presented pri-orities the following topics can be investigated in future studies

(1) Presenting a comprehensive model for generatingvarious cryptography algorithms based on the pri-orities of interest

(2) Examining and formulating a model about the roleof basic science for design and implementation ofcryptography algorithms considering the significantrole of ldquobasic sciencerdquo in implementation of thecryptography algorithms and about the role ofschools and universities

(3) Presenting a model for design and implementationof cryptography algorithms in the IoT infrastructurewith optimal resource usage

Data Availability

e data used to support the findings of the study areavailable from the corresponding author upon request

Conflicts of Interest

e authors declare that there are no conflicts of interestregarding the publication of this article

Table 15 Irregularity of the cryptography algorithm implementation techniques for α3 24

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d15 qi 2 1 3 2 1- qi 1 1 2 033

11117

1 1 1

091

1112

com

plex

ity

alpha-cutα=2 α=3 α=5 α=15 α=25

Figure 12 System complexity of cryptography algorithm implementation techniques for different alpha-cuts

Table 16 Ranking the implementation techniques of the cryptography algorithms considering various alpha-cuts using Q-analysis

Equivalence classes for the parameters with minimum relationship level (q 0) αBasic sciences α20 163Key α10 81Hardware and steganography α5 40Avalanche effect and hybrid method α3 24Artificial intelligence software occupation area energy consumption memory consumption α2 16

basic sciences

key

Avalanche effect hybrid method

hardware steganography

artificial intelligence soware occupationarea energy consumption memory consumption

Figure 13 Prioritizing the implementation techniques of thecryptography algorithms using Q-analysis

Computational Intelligence and Neuroscience 15

Acknowledgments

is work was partially supported by Iran National ScienceFoundation (INSF) under contract No 9653979

References

[1] D Nilesh and M Nagle ldquoe new cryptography algorithmwith high throughputrdquo in Proceedings of the 2014 Interna-tional Conference on Computer Communication and Infor-matics pp 3ndash7 Coimbatore India January 2014

[2] S Ravi P Kocher R Lee G McGraw and A RaghunathanldquoSecurity as a new dimension in embedded system designrdquo inProceedings of the 41st annual conference on Design auto-mation-DAC lsquo04 pp 753ndash760 San Diego CA USA March2004

[3] P Kuppuswamy and S Q Y A Khalidi ldquoHybrid encryptiondecryption technique using new public key and symmetric keyalgorithmrdquo International Journal of Information and Com-puter Security vol 6 no 4 pp 372ndash382 2014

[4] S Bhat and V Kapoor ldquoSecure and efficient data privacyauthentication and integrity schemes using hybrid cryptog-raphyrdquo International Conference on Advanced ComputingNetworking and Informatics vol 870 pp 279ndash285 2019

[5] D Ott and C Peikert ldquoAnd other workshop participantsldquoidentifying research challenges in post quantum cryptog-raphy migration and cryptographic agilityrdquo 2019 httparxivorgabs190907353

[6] P Manickam K Shankar E Perumal M Ilayaraja andK Sathesh Kumar ldquoSecure data transmission through reliablevehicles in VANET using optimal lightweight cryptographyrdquoin Advanced Sciences and Technologies for Security Applica-tions pp 193ndash204 Springer New York NY USA 2019

[7] S Prakash and A Rajput ldquoHybrid cryptography for securedata communication in wireless sensor networksrdquo Advancesin Intelligent Systems and Computing vol 696 pp 589ndash5992018

[8] A A Soofi I Riaz and U Rasheed ldquoAn enhanced vigenerecipher for data securityrdquo International Journal of Scientific ampTechnology Research vol 4 no 8 pp 141ndash145 2015

[9] N Shamsuddin M Syafiqah and S Ali Pitchay ldquoImple-menting location-based cryptography on mobile applicationdesign to secure data in cloud storagerdquo Journal of PhysicsConference Series IOP Publishing vol 1551 no 1 2020

[10] L Jiao H Yonglin and D Feng ldquoStream cipher designs areviewrdquo Science China Information Sciences vol 63 no 3pp 1ndash25 2020

[11] NIST ldquoNIST cryptographic standards and guidelines devel-opment processrdquo Inside NIST vol 27 2016

[12] G Alagic J Alperin-Sheriff D Apon et al Status Report onthe First Round of the NIST Post-Quantum CryptographyStandardization Process National Institute of Standards andTechnology Gaithersburg MD USA pp 1ndash27 2019

[13] I Damaj and S Kasbah ldquoAn analysis framework for hardwareand software implementations with applications from cryp-tographyrdquo Computers amp Electrical Engineering vol 69pp 572ndash584 2018

[14] M S Croock Z A Hassan and S D Khudhur ldquoAdaptive keygeneration algorithm based on software engineering meth-odologyrdquo International Journal of Electrical and ComputerEngineering vol 11 no 1 p 589 2021

[15] K Sha T A Yang W Wei and S Davari ldquoA survey of edgecomputing-based designs for iot securityrdquo Digital Commu-nications and Networks vol 6 no 2 pp 195ndash202 2020

[16] M Yahuza M Y I B Idris A W B A Wahab et alldquoSystematic review on security and privacy requirements inedge computing state of the art and future research oppor-tunitiesrdquo IEEE Access vol 8 pp 76541ndash76567 2020

[17] A Bhowmik J Dey and S Karforma A Novel Double TierCryptographic System (nDTCS) to Reinforce Patientsrsquo Privacyin Contemporary COVID-19 Telemedicine Springer NewYork NY USA 2021

[18] N Gupta User Integrity Protection Security Model forEnhanced Telemedicine for Healthcare Networks 2020httpsosfio69jdq

[19] J Dey A Bhowmik A Sarkar S Karforma andB Chowdhury Cryptographic Engineering on COVID-19Telemedicine An Intelligent Transmission through RecurrentRelation Based Session Key Springer New York NY USA2021

[20] V Hosseinian H Ayatollahi H Haghani and E MehraeenldquoRequirements of information security in a telemedicinenetwork review of IT managersrsquo opinionrdquo Journal of Para-medical Sciences amp Rehabilitation vol 4 no 2 pp 31ndash402015

[21] D Karaoglan Altop A Levi and V Tuzcu ldquoDeriving cryp-tographic keys from physiological signalsrdquo Pervasive andMobile Computing vol 39 pp 65ndash79 2017

[22] O Uzunkol and M S Kiraz ldquoStill wrong use of pairings incryptographyrdquo Applied Mathematics and Computationvol 333 pp 467ndash479 2018

[23] S Picek D Jakobovic J F Miller L Batina and M CupicldquoCryptographic Boolean functions one output many designcriteriardquo Applied Soft Computing vol 40 pp 635ndash653 2016

[24] R Saha and G Geetha ldquoSymmetric random function gen-erator (SRFG) a novel cryptographic primitive for designingfast and robust algorithmsrdquo Chaos Solitons amp Fractalsvol 104 pp 371ndash377 2017

[25] A Achuthshankar and A Achuthshankar ldquoA novel sym-metric cryptography algorithm for fast and secure encryp-tionrdquo in Proceedings of the 2015 IEEE 9th InternationalConference on Intelligent Systems and Control (ISCO) pp 4ndash9Coimbatore India January 2015

[26] D Ott and C Peikert ldquoIdentifying research challenges in postquantum cryptography migration and cryptographic agilityrdquo2019 httparxivorgabs190907353

[27] W Liu B Ying H Yang and H Wang ldquoAccurate modelingfor predicting cryptography overheads on wireless sensornodesrdquovol 2 pp 997ndash1001 in Proceedings of the 2019 11thInternational Conference on Advanced CommunicationTechnology vol 2 pp 997ndash1001 IEEE Bangalore IndiaFebruary 2019

[28] S Feizi A Ahmadi and A Nemati ldquoA hardware imple-mentation of simon cryptography algorithmrdquo in Proceedingsof the 2014 4th International Conference on Computer andKnowledge Engineering (ICCKE) pp 245ndash250 IEEEMashhad Iran 2014

[29] E ambiraja G Ramesh and R Umarani ldquoA survey onvarious most common encryption techniquesrdquo 2012 httpwwwijarcssecom

[30] A Gupta and N K Walia ldquoCryptography Algorithms areviewrdquo International Journal of Engineering Research andDevelopment vol 2 no 2 pp 1667ndash1672 2014

[31] R Bhanot and R Hans ldquoA review and comparative analysis ofvarious encryption algorithmsrdquo International Journal of Se-curity and Its Applications vol 9 no 4 pp 289ndash306 2015

[32] S Chandra S Paira S S Alam and G Sanyal ldquoA com-parative survey of symmetric and asymmetric key

16 Computational Intelligence and Neuroscience

cryptographyrdquo in Proceedings of the 2014 InternationalConference on Electronics Communication and Computa-tional Engineering (ICECCE) pp 83ndash93 Hosur India No-vember 2014

[33] F Maqsood M Ahmed M Mumtaz and M Ali ldquoCryp-tography a comparative analysis for modern techniquesrdquoInternational Journal of Advanced Computer Science andApplications vol 8 no 6 p 6 2017

[34] S B Sadkhan and A O Salman ldquoA survey on lightweight-cryptography status and future challengesrdquo in Proceedings ofthe International Conference on Advances in SustainableEngineering and Applications pp 105ndash108 Wasit Iraq June2018

[35] N V oai ldquoCriteria and dimension reduction of linearmultiple criteria optimization problemsrdquo Journal of GlobalOptimization vol 52 no 3 pp 499ndash508 2012

[36] C E Shannon ldquoCommunication theory of secrecy systems1945rdquo MD Computing vol 15 no 1 pp 57ndash64 1998

[37] L J Fennelly M Beaudry and M A Perry Security in 2025[38] ITU Global Cybersecurity Index 2018 2019[39] ITU Global Cybersecurity Index (GCI) 2017 2017

Computational Intelligence and Neuroscience 17

(i) e relationship matrix of support indices withmain design and implementation processes of thecipher algorithms since there are 13 support indicesin the level-3 model (Figure 3) and 4 steps in thelevel-1 model (Figure 2) a 13lowast4 matrix is consti-tuted to determine the relationship between themembers of these two processes where its rows arethe elements of the support process and its columnsare the quadruple elements of the design andimplementation of cryptography algorithms eelements of this matrix are between 0 and 10 asshown in Table 2 [35] e value of each elementrepresents the effectiveness of each support index oneach design and implementation step Analysis ofthis matrix and its modeling computation providesthe possibility for the policy-makers and developersto manage the resources required for each step ofdesign and implementation of the cryptographyalgorithms and use the available hardware andsoftware resources optimally

(ii) e relationship matrix of the objectives with themain design process of the cryptography algo-rithms considering the seven objectives extracted in

Section 2 and four steps in the main process todetermine the role of each step in achieving theseven objectives a 7lowast4 matrix is constituted Sincethe number of final resources used to extract data is814 papers technical reports and documents theelements of this matrix are between 0 and 814 asgiven in Table 2 In this matrix the value of eachelement represents the number of studies docu-ments and reports indicating the relationship be-tween two components An interesting point in thismatrix is that all algorithms implemented in thestudied documents are evaluated

(iii) e relationship matrix of the implementationtechniques and the design objectives of the cryp-tography algorithms considering the 29 extractedtechniques for cryptography algorithm imple-mentation and seven main objectives of the cryp-tography algorithms a 29lowast 7 matrix is constitutedto determine how much each technique is used forcryptography implementation to achieve each ob-jective (Table 3) e elements of this matrix arebetween 0 and 814 A part of this matrix is shown inTable 4

ICT

RampD

Financial

Rule amp Regulation

Standard

Human resourceEducation

OrganizationStructure

Management

Cultural

International relation

Infrastructure

Material

equipments

Hash Function

ControllerProcess

Product

Policy

Requ

est

Plai

n Te

xt

Cipher Text

Supportive

Main Processes

Software

Symmetric Algorithm

Asymmetric Algorithm

Hardware

Figure 3 e cybernetic model of design and implementation of cipher algorithms at level 3

Computational Intelligence and Neuroscience 5

311 Design of Cryptography Algorithms Using the Cyber-netic Supermodel Considering the above discussion andpresence of numerous indices and components in thedesign and implementation of the cryptography algorithmwhich make it a complex system identifying the rela-tionship between these indices and ranking them is anecessity Accordingly in this section the Q-analysis

method is used and the output of the three matrices isanalyzed According to the level three of the proposedcybernetic model the design and implementation supportprocess of the cryptography algorithms includes 13 com-ponents On the contrary the design and implementationsteps of the cryptography algorithms also include foursteps constituting a 13 lowast 4 matrix

Table 2 e 13-component interaction matrix of the support and the main design and implementation processes of cryptographyalgorithms

Main process Level 1Cryptography algorithms Level 2

Application eoretical basis Implementation Evaluation Level 3

Support process

Software

Culture 5 6 5 2 Organizationstructure 8 7 7 7

International and public relations 5 4 4 7 Financial resources 7 6 7 6

Human resources and education 8 10 10 9 Research and development 10 8 8 10

Rules 7 4 5 10 Management 9 7 8 7

FAVA 10 5 8 7 Standard 10 3 8 10

HardwareEquipment 8 6 7 8

Infrastructure 9 5 5 7Material 7 2 5 5

Table 3 e relationship matrix of the objectives with the design process of the cryptography algorithms

Main process objectives Application eoretical basis Implementation EvaluationSecurity 203 386 516 516Simplicity 5 23 38 38Resources 41 30 81 81Flexibility 15 28 45 45Scalability 6 11 22 22Speed 72 146 247 247Reliability 2 10 11 11

Table 4 A part of the interaction matrix between the cryptography algorithm implementation techniques and the seven objectives

Level 3 Security Simplicity Using resources Flexibility Scalability Speed Reliability

Support process

Avalanche effect 32 0 0 0 0 0 0Digital signature 1 0 0 0 0 0 0

Block size 1 1 0 0 2 0 0Image sharing 0 0 0 1 0 0 0

Parallel processing 0 0 0 1 12 0 0reshold technique 3 1 0 0 0 0 0

Data mining 1 0 0 0 0 0 0Binary tree 1 0 0 0 3 0 0

Cycle 5 2 0 0 12 0 0Multistage crypto 1 0 0 0 0 0 0Hybrid method 30 3 0 2 9 0 2

Hardware 0 0 7 0 0 0 0

6 Computational Intelligence and Neuroscience

32 Calculating the Incidence Matrix First the incidencematric is obtained based on the data matrix 4ndash1 is matrixrepresents the ldquoimpact of support indices on the main designprocess of the cryptography algorithmsrdquo e data matrix iscomprised of two sets D support indices set C and qua-druple design and implementation steps (Table 5) e in-cidence matrix calculated using the data matrix for α 70is represented in Table 6 By assigning different values to theparameter α difference incidence matrices are obtainederesults of the Q-analysis using C++ coding for α 70 aregiven in Figure 4

D d1 d2 d131113864 1113865

C c1 c2 c3 c41113864 1113865(1)

321 Geometric Representation Multidimensional proper-ties of the system are defined by a simple or complex setKD(C λ) such that the entities of the set D represent thesupport indices and entities of the set C represent thequadruple design and implementation steps of the cryp-tography algorithms

In the sample with α70 7 dis is as follows

d1

d2 c1 c2 c3 c41113864 1113865

d3 c41113864 1113865

d4 c1 c31113864 1113865

d5 c1 c2 c3 c41113864 1113865

d6 c1 c2 c3 c41113864 1113865

d7 c1 c41113864 1113865

d8 c1 c2 c3 c41113864 1113865

d9 c1 c3 c41113864 1113865

d10 c1 c3 c41113864 1113865

d11 c1 c3 c41113864 1113865

d12 c1 c41113864 1113865

(2)

e simplexes of σp(di) are also

σminus1 d1( 1113857 σ3 d2( 1113857 σ0 d3( 1113857 σ1 d4( 1113857 σ3 d5( 1113857 σ0 d3( 1113857 σ1 d7( 1113857

σ3 d8( 1113857 σ2 d9( 1113857 σ2 d10( 1113857 σ2 d11( 1113857 σ1 d12( 1113857 σ1 d13( 1113857 (3)

erefore the complex dimension is 3 In other wordsthe diagnosis classes d2 (structureorganization) d5 (humanresourceseducation) d6 (research and development) andd8 (management) have the largest dimension

33 CalculatingDimensions andQ-Link Q-link is defined asthe link between a subset with smallest interface between twosubsequent dis in the chain of d1 to dn Q-link between twosubsequent dis with α70 7 is

Table 5 Sets of dis and cis support indices based on the datamatrix

d1 Culture

d2 Structureorganizationd3 Internationalpublic relationsd4 Financiald5 Human resourceseducationd6 Research and developmentd7 Rulesd8 Managementd9 FAVAd10 Standardd11 Equipmentd12 Infrastructured13 MaterialC1 ApplicationC2 eoretical basisC3 ImplementationC4 Evaluation

Table 6 e incidence matrix of the support indicesrsquo impact of thedesign steps of the cryptography algorithms with α 70

C1 C2 C3 C4

d1 0 0 0 0d2 1 1 1 1d3 0 0 0 1d4 1 0 1 0d5 1 1 1 1d6 1 1 1 1d7 1 0 0 1d8 1 1 1 1d9 1 0 1 1d10 1 0 1 1d11 1 0 1 1d12 1 0 0 1d13 1 0 0 0

Computational Intelligence and Neuroscience 7

σminus1 d1( 1113857 σ3 d2( 1113857⟶ 1 σ3 d2( 1113857 σ0 d3( 1113857⟶ 0 σ0 d3( 1113857 σ1 d4( 1113857⟶ minus 1

σ1 d4( 1113857 σ3 d5( 1113857⟶ 1 σ3 d5( 1113857 σ3 d6( 1113857⟶ 3 σ3 d6( 1113857 σ1 d7( 1113857⟶ 1

σ1 d7( 1113857 σ3 d8( 1113857⟶ 1 σ3 d8( 1113857 σ2 d9( 1113857⟶ 2 σ2 d9( 1113857 σ2 d10( 1113857⟶ 2

σ2 d10( 1113857 σ2 d11( 1113857⟶ 2 σ3 d11( 1113857 σ1 d12( 1113857⟶ 1 σ1 d12( 1113857 σ0 d13( 1113857⟶ 0

(4)

e maximum link dimension is 3 indicating the re-lationship between the diagnosis classes

331 Calculating the Structure Vectors As mentioned inthe definitions the vector Qq is a simplification basiscreated to eliminate the additional impacts in the equiv-alent simplex sets e maximum complex dimension withα70 7 is 3 erefore the first structure vector based onthe output is

Dimension 3 2 1 0 (5)

e second structure vector P is

dimensions 3 2 1 0

P Pdim3 Pdim2 Pdim1 Pdim0( 1113857

P 4 7 10 12( 1113857

(6)

where Pq is the number simplexes greater than or equal to qin the set K in which P is the number of simplex linkrepetitions (support indices) in the quadruple design andimplementation steps of the cryptography algorithms Basedon the values of these two vectors it is seen that the rela-tionship between the support indices and the quadrupledesign and implementation steps of the cryptography al-gorithms is high is issue indicates the role of supportcomponents in the design and implementation of the

cryptography algorithms which should be consideredseriously

332 Calculating the Obstruction or Flexibility VectorQlowastK represents the number of structural obstructions forsimplex interactions in dimension k

Qlowast

Q minus I⟶ Qlowast

1 1 1 11113858 1113859 minus 1 1 1 11113858 1113859⟶ Qlowast

0 0 0 01113858 1113859

(7)

As can be seen there is no obstruction in any of thecommunication levels indicating that there is a significantrelationship between the support components at eachequivalence class

333 Calculating Irregularity e value of (eccprime(σ)) iscalculated using the Chinese method e results for α70

7 are shown in Table 7 e calculated irregularity valueshows that the indices d2 (structureorganization) d5 (hu-man resourceseducation) d6 (research and development)and d8 (management) affect other indices

334 Calculating Complexity Also the results ofQ-analysiscan be used to describe structure complexity According toequations (4)ndash(9) and for α = 7 the complexity measure is

-1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1

3 0 1 3 3 1 3 2 2 2 1

0 -1 0 0 0 0 0 0 0 0

1 1 1 0 1 1 1 1 0

3 3 1 3 2 2 2 1

3 1 3 2 2 2 1

3 2 2 2 1

2 2 2 1

2 2 1

2 1

1

1 1 1 1 1 1

-1

-1

0

0

0

0

0

0

0

0

0

0

0q Q3210

11

11

SETSX2 X5 X6 X8X2 X5 X6 X8 X9 X10 X11X2 X4 X5 X6 X7 X8 X9 X10 X11 X12X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13

Figure 4 Implementation results of the model for support components of the cryptography algorithmsrsquo design with α 70

8 Computational Intelligence and Neuroscience

Q 1 1 1 1( 1113857

ψ(K) 2(1 + 2 + 3 + 4)

(4lowast 5)1113890 1113891 1

(8)

Since in the above model there is no obstruction amongthe component of the equivalent class it was expected thatthe complexity of the support components is not high andthe obtained complexity index of 1 verifies this expectatione system complexity for different alpha-cuts is shown inFigure 5

335 Ranking the Support Components of the Design andImplementation of the Cryptography Algorithms e resultsof using A-analysis are shown in Table 7 e connectionstrength of the factors in one group is specified with alpha-cut erefore the support components are grouped in 5levels Each level describes the priority and importance of thegroup in developing the cryptography algorithms In Fig-ure 6 the ranking pyramid of the support components usingQ-analysis is shown To allocate proper resources thecomponents existing in higher levels of the pyramid (Fig-ure 6) are of higher priority

336 Validation of the Results In this section the resultsof the cybernetic model and Q-analysis for supportcomponentsrsquo ranking are compared with the results re-ported in the global cybersecurity index (GCI) in 20152017 and 2018 presented by ITU [36ndash39] e GCI reportsare focused on five indices including ldquolegal cases orga-nization necessities technical issues capacity buildingand cooperationrdquo and the subindices include legaltechnical organization capacity building and coopera-tion According to the presented indices and subindices itis clear that the ldquorulesrdquo ldquostandardrdquo ldquoresearch and de-velopmentrdquo ldquoeducationrdquo and ldquomanagementrdquo are ofhigher priority in security establishment Although ourresearch is more skilled and detailed compared to the GCIreports but the results verify our findings Also Table 8shows the ranking of the support components obtainedusing Q-analysis

337 Executing the Model and Analyzing the Results of theObjectivesrsquo Impact on Design Steps of the Cryptography Al-gorithmsrsquo Matrix In this section the role of the sevencomponents on the quadruple design and implementationsteps of the cryptography algorithm is analyzed with Q-analysis Using the Q-analysis method and the 7lowast4 matrixobtained from the relationship of the cryptography algo-rithmsrsquo design objectives on their quadruple steps theirindices are ranked

338 Calculating the Incidence Matrix and Executing theModel First the incidence matrix is obtained for the datamatrix shown in Figure 2 It can be seen in Table 9 that eachelement of the two sets represents which indice e inci-dence matrix calculated from the data matrix for α5 40 is

given in Table 10 e results of implementing the model forα5 40 are given in Figure 7

4 Geometric Representation

In the sample with α5 40 dis are

d1 c1 c2 c3 c41113864 1113865

d2

d3 c1 c3 c41113864 1113865

d4 c3 c41113864 1113865

d5

d6 c1 c2 c3 c41113864 1113865

d7

(9)

e simplexes of σp(di) are also

σ3 d1( 1113857σminus1 d2( 1113857σ2 d3( 1113857σ1 d4( 1113857σminus1 d5( 1113857σ3 d6( 1113857σminus1 d7( 1113857

(10)

erefore the complex dimension is 3 In other wordsthe discriminant classes d1 (security) and d6 (speed) have thelargest dimension

41 Calculating the Structure Vectors As mentioned thevector Qq is a simplification basis to eliminate the additionaleffects in the set of equivalent simplexes e maximumcomplex dimension for α5 40 is 3 erefore the firststructure vector based on the output is

Dimension 3 2 1 1

Q 1 1 1 1( 1113857

Dimension 3 2 1 0

P 2 3 4 7( 1113857

(11)

e second structure vector P is

P 2 3 4 7( 1113857 (12)

42 Calculating the Obstruction Vector or InflexibilityQlowastK represents the number of structural obstructions forsimplex interactions in dimension k which is as follows forα5 40

1 1 1 1 1332

005

115

225

com

plex

ity

alpha cutα=5 α=6 α=7 α=8 α=9 α=10

Figure 5 System complexity of the support indices of design andimplementation of the cryptography algorithms for different alpha-cuts

Computational Intelligence and Neuroscience 9

Qlowast

Q minus I⟶ Qlowast

2 1 1 31113858 1113859 minus 1 1 1 11113858 1113859⟶ Qlowast

0 0 0 01113858 1113859

(13)

According to the obtained values it can be concludedthat there are no structural obstructions among the main

design indices of the cryptography algorithms at ismultiple objectives are considered simultaneously by thecryptography algorithm designers

Table 7 e irregularity of the data matrix parameters for α= 7

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d2 d5 d6 d8 qi 3 2 1 364 3 061d9 d10d11 qi 2 1 064 3 011d4 d7 d12 qi 1 014 3 002

human resourceRampD

rules FAVA standard

management infrastructure

organizationstructure equipmentinternational and public relations finanacial

resources and material

Figure 6 e ranking pyramid of the support components using Q-analysis

Table 8 Ranking of the support components using Q-analysis

Relationship of the support components with the design and implementation of the cryptography algorithms (q 0)No relationship α 0 complete relationship α 100

Human resource RampD rules FAVA and standard α100 10Management and infrastructure α90 9Organizationstructure and equipment α80 8International and public relations financial resources and material α70 7Culture α60 6All components α60 5

Table 9 Set of dis and cis the objectivesrsquo indices

eoretical basis C2 Reliability d7 Scalability d5 Resources d3 Security d1Implementation C3 Application C1 Speed d6 Flexibility d4 Simplicity d2Evaluation C4

Table 10 e incidence matrix of the design objectives of thecryptography algorithms with α= 40

c1 c2 c3 c4

d1 1 1 1 1d2 0 0 0 0d3 1 0 1 1d4 0 0 1 1d5 0 0 0 0d6 1 1 1 1d7 0 0 0 0

-1 -1 -1 -1 -1

2 1 -1 2

1 -1 1

-1 -1

3

3 -1 2 1 -1 3 -1

-1

-1

-1

-1

-1

-1q Q3210

11

11

SETSX1X6X1X3X6X1X3X4X6X1X3X4X6

Figure 7 e results of executing the model for design objectivesrsquoindices with α 40

10 Computational Intelligence and Neuroscience

43 Calculating Irregularity e results of applying theChinese method (ecclsquo(σ)) for calculating irregularity forα5 40 are shown in Table 11 e calculated irregu-larity value shows that indices d1 (security) and d6(speed) have received more attention compared to otherindices

44 Calculating Complexity Structural complexity for asample with α5 40 is

dimensions 3 2 1 0

Q Qdim3 Qdim2 Qdim1 Qdim0( 1113857

Q 1 1 1 1( 1113857

ψ(K) 2(1 + 2 + 3 + 4)

(4lowast 5)1113890 1113891 1

(14)

Since in the above model there was no obstructionamong the components of the equivalent classes in any of thecommunication levels it was expected that there is not ahigh complexity among objectives of the cryptography al-gorithm design and the complexity index of 1 verified thisexpectation e system complexity for different alpha-cutsis shown in Figure 8 As can be seen the system complexity is1 for all significant values of alpha

45 Prioritizing the Parameters of the Main Objectives of theCryptography Algorithm Design Using Q-AnalysisAccording to the analysis results the parameters can beclassified into 5 levelsese levels are shown in Table 12 andFigure 9 Each level indicates a priority and importance ofthe group in the development of cryptography algorithmsTo allocate proper resources the components at the higherlevels of the pyramid are of higher priority As can be seenthree objectives of security speed and optimal usage ofresources have the highest priority for the design of cryp-tography algorithms

46 Executing the Model and Analyzing the Results for theCryptography Algorithm Implementation Techniques einteraction matrix between the 29 extracted techniques andthe sevenmain objectives is shown in Figure 10e purposeof this section is to rank the cryptography algorithmimplementation techniques to achieve the goals of interest

47Calculating the IncidenceMatrixandExecuting theModele data matrix A is comprised of two sets e set Drepresents the employed techniques and the set C representsthe seven main design objectives (Table 13)

D d1 d2 d291113864 1113865

C c1 c2 c71113864 1113865(15)

Some parts of the incidence matrix calculated from thedata matrix A for α3 are shown in Table 14 and the resultsof theQ-analysis model for α3 24 are shown in Figure 11

48 Geometric Representation In the sample with α3 24the dis are

d1 c11113864 1113865

d2

d3

d4

d5

d6

d7

d8

d9

d10

d11 c11113864 1113865

d12

d13 c51113864 1113865

d14

d15 c1 c4 c51113864 1113865

d16

d17

d18

d19 c11113864 1113865

d20

d21

d22 c31113864 1113865

d23

d24

d25

d26

d27 c11113864 1113865

d28

d29

(16)

e simplexes of σp(di) are

Computational Intelligence and Neuroscience 11

σ0 d1( 1113857

σminus1 d2( 1113857

σminus1 d3( 1113857

σminus1 d4( 1113857

σminus1 d5( 1113857

σminus1 d6( 1113857

σminus1 d7( 1113857

σminus1 d8( 1113857

σminus1 d9( 1113857

σminus1 d10( 1113857

σ0 d11( 1113857

σminus1 d12( 1113857

σ0 d13( 1113857

σminus1 d14( 1113857

σ2 d15( 1113857

σminus1 d16( 1113857

σminus1 d17( 1113857

σminus1 d18( 1113857

σ0 d19( 1113857

σminus1 d20( 1113857

σminus1 d21( 1113857

σ0 d22( 1113857

σminus1 d23( 1113857

σminus1 d24( 1113857

σminus1 d25( 1113857

σminus1 d26( 1113857

σ0 d27( 1113857

σminus1 d28( 1113857

σminus1 d29( 1113857

(17)

erefore the complex dimension is 2 In other wordsthe discriminant class d15 (basic sciences) has the largestdimension

49 Calculating the Dimensions and the Q-Link In the alphadefined for cryptography algorithm implementation tech-niques to achieve the defined goals the obtained q-linkshows that these techniques are relatively independent al-though there is a weak relationship between some tech-niques Q-link in the samples with α3 24 between eachtwo subsequent di is as follows

σ0 d1( 1113857 σminus1 d2( 1113857⟶ minus 1

σminus1 d2( 1113857 σminus1 d3( 1113857⟶ minus 1

σ0 d3( 1113857 σminus1 d4( 1113857⟶ minus 1

σminus1 d4( 1113857 σminus1 d5( 1113857⟶ minus 1

σminus1 d5( 1113857 σminus1 d6( 1113857⟶ minus 1

σminus1 d6( 1113857 σminus1 d7( 1113857⟶ minus 1

σminus1 d7( 1113857 σminus1 d8( 1113857⟶ minus 1

σminus1 d8( 1113857 σminus1 d9( 1113857⟶ minus 1

σminus1 d9( 1113857 σminus1 d10( 1113857⟶ minus 1

σminus1 d10( 1113857 σ0 d11( 1113857⟶ minus 1

σ0 d11( 1113857 σminus1 d12( 1113857⟶ minus 1

σminus1 d12( 1113857 σ0 d13( 1113857⟶ minus 1

σ0 d13( 1113857 σminus1 d14( 1113857⟶ minus 1

σminus1 d14( 1113857 σ2 d15( 1113857⟶ minus 1

σ2 d15( 1113857 σminus1 d16( 1113857⟶ minus 1

σminus1 d16( 1113857 σminus1 d17( 1113857⟶ minus 1

σminus1 d17( 1113857 σminus1 d18( 1113857⟶ minus 1

σminus1 d18( 1113857 σ0 d19( 1113857⟶ minus 1

σ0 d19( 1113857 σminus1 d20( 1113857⟶ minus 1

σminus1 d20( 1113857 σminus1 d21( 1113857⟶ minus 1

σminus1 d21( 1113857 σ0 d22( 1113857⟶ minus 1

σ0 d22( 1113857 σminus1 d23( 1113857⟶ minus 1

σminus1 d23( 1113857 σminus1 d24( 1113857⟶ minus 1

σminus1 d24( 1113857 σminus1 d25( 1113857⟶ minus 1

σminus1 d25( 1113857 σminus1 d26( 1113857⟶ minus 1

σminus1 d26( 1113857 σ0 d27( 1113857⟶ minus 1

σ0 d27( 1113857 σminus1 d28( 1113857⟶ minus 1

σminus1 d28( 1113857 σminus1 d29( 1113857⟶ minus 1

(18)

Table 11 Irregularity of the cryptography objectivesrsquo parameters in the data matrix A for α5

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d1 and d6 qi 3 2 and 1 433 3 072d3 qi 2 and 1 133 3 026d4 qi 1 033 3 006

11117

1 1 1

09095

1105

11115

12

com

plex

ity

alpha cutα=2 α=3 α=5 α=15 α=25

Figure 8 System complexity of the cryptography algorithm designobjectives for different alpha-cuts

12 Computational Intelligence and Neuroscience

410 Calculating the Structure Vectors erefore the firststructure vector based on the software output is

Q 1 1 2( 1113857 (19)

e second structure vector P is

P 1 1 29( 1113857 (20)

Since the large values of P of higher dimensions dem-onstrate more links the second structure vector calculatedfor the alpha of interest shows that the relationship betweenthe cryptography algorithm implementation techniques isminimum

411 Calculating the Obstruction or Inflexibility Vectore obstruction vector (Qlowast) for a sample with α3 24 is

Table 12 Ranking the objectivesrsquo parameters considering different cuts in Q-analysis

e equivalence class for the parameters with minimum relation level (q 0) αSecurity α50 244Speed α30 244Resources α15 122Flexibility α5 40Simplicity-scalability α2 16(All parameters)

securityspeed

optimal usage of resources

flexibility

simplicity-scalability

Figure 9 Prioritizing the design and implementation objectives of the cryptography algorithms using Q-analysis

securityspeed

optimal usage of resources

flexibility

simplicity-scalability-reliability

Figure 10 Prioritizing the design and implementation objectives of the cryptography algorithms using Q-analysis

Table 13 Set of dis and cis the techniques employed for implementing cryptography algorithms

Flexibility C4 Fuzzy logic d25 Storage space d17 Cycle d9 Avalanche effect d1Scalability C5 Software d26 Clustering d18 Multiple step d10 Digital signature d2Speed C6 Steganography d27 Key d19 Hybrid method d11 Block size d3Reliability C7 Music harmony d28 Graph d20 Hardware d12 Image sharing d4

Artificial intelligence d29 Characteristic oriented d21 Hardware d13 Parallel processing d5Security C1 Energy consumption d22 Occupation area d14 reshold technique d6Simplicity C2 Bandwidth consumption d23 Basic science d15 Data mining d7Resources C3 Memory consumption d24 Compression d16 Binary tree d8

Table 14 e incidence matrix of the cryptography algorithmimplementation techniques with α 24

C1 C2 C3 C4 C5 C6 C7

d1 1 0 0 0 0 0 0d2 0 0 0 0 0 0 0d3 0 0 0 0 0 0 0d4 0 0 0 0 0 0 0d5 0 0 0 0 0 0 0d6 0 0 0 0 0 0 0d7 0 0 0 0 0 0 0d8 0 0 0 0 0 0 0d9 0 0 0 0 0 0 0d10 0 0 0 0 0 0 0d11 1 0 0 0 0 0 0d12 0 0 0 0 0 0 0

Computational Intelligence and Neuroscience 13

Qlowast

Q minus I⟶ Qlowast

1 1 21113858 1113859 minus 1 1 11113858 1113859⟶ Qlowast

0 0 11113858 1113859

(21)

e value of QlowastK represents the number of structurallimitations or obstructions for the cryptography techniquesrsquointeraction at dimension k As can be seen from the cal-culated obstruction vector there is a significant relationshipat some levels

412 Calculating Irregularity Irregularity is the integrationdegree of a cryptography method in the total complex Mea-suring irregularity (eccrsquo) for α3 24 is shown in Table 15 Ascan be seen irregularity for the discriminant class d15 indicatesthat this technique is isolated from other techniques

413 Calculating Complexity Structure complexity for asample with α3 24 is

Q 1 1 2( 1113857

ψ(K) 2(2 + 2 + 3)

(3lowast 4)1113890 1113891 117

(22)

e system complexity for different values of alpha isshown in Figure 12 As can be seen the total complexity ofthe system for large values alpha tends to stability

414 Ranking the Cryptography Algorithm ImplementationTechniques ese techniques can be classified into 5 levelsusing the analysis of the obtained results ese levels can beseen in Table 16 and Figure 13 e link power of the factorsin one group is specified with alpha Each level describespriority and importance of the group in the development ofcryptography algorithms Accordingly six methods of basicscience key management using hardware methods usingsteganography Avalanche effect and using hybrid method

q Q21o

112

SETSX15X15X1X11X15X19X27X13 amp X22

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-10-1-1-10-1-1-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-10-1-1-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

2-10-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1

-1-1-1-1-1-1

-1-1-1-1-1

-1-1-1-1

-1-1-1

-1-10

Figure 11 Results of executing the model for cryptography algorithm implementation techniques with α3 24

14 Computational Intelligence and Neuroscience

are the most important cryptography algorithm imple-mentation techniques to achieve the main objectives

5 Conclusion

In this study the design and implementation model of thecryptography algorithms is designed in three levels and itseffective components are extracted To organize the com-ponents in addition to elimination of the additional com-ponents through the measure reduction algorithm a properclassification is applied to examine the mutual effects Afterclassification three 13lowast 4 7lowast 4 and 29lowast 7 matrices areconstituted and the model is implemented on these threematrices To implement the designed model Q-analysis isused Accordingly for support indices five indices of highpriority include human resources research and develop-ment management organizationstructure and equipmentFor the algorithm design objectives index five high priorityindices include security speed optimal usage of resources

and simplicity For the indices related to implementationtechniques of the cryptography algorithms the most appliedtechniques for achieving the determined objectives includeusing basic science hardware and software methods usingkeymanagement hybrid method steganography Avalancheeffect and using artificial intelligence In the future work wewill consider the following cases

A plan should be formulated according to the prioritiesAccording to the outputs of this study and the presented pri-orities the following topics can be investigated in future studies

(1) Presenting a comprehensive model for generatingvarious cryptography algorithms based on the pri-orities of interest

(2) Examining and formulating a model about the roleof basic science for design and implementation ofcryptography algorithms considering the significantrole of ldquobasic sciencerdquo in implementation of thecryptography algorithms and about the role ofschools and universities

(3) Presenting a model for design and implementationof cryptography algorithms in the IoT infrastructurewith optimal resource usage

Data Availability

e data used to support the findings of the study areavailable from the corresponding author upon request

Conflicts of Interest

e authors declare that there are no conflicts of interestregarding the publication of this article

Table 15 Irregularity of the cryptography algorithm implementation techniques for α3 24

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d15 qi 2 1 3 2 1- qi 1 1 2 033

11117

1 1 1

091

1112

com

plex

ity

alpha-cutα=2 α=3 α=5 α=15 α=25

Figure 12 System complexity of cryptography algorithm implementation techniques for different alpha-cuts

Table 16 Ranking the implementation techniques of the cryptography algorithms considering various alpha-cuts using Q-analysis

Equivalence classes for the parameters with minimum relationship level (q 0) αBasic sciences α20 163Key α10 81Hardware and steganography α5 40Avalanche effect and hybrid method α3 24Artificial intelligence software occupation area energy consumption memory consumption α2 16

basic sciences

key

Avalanche effect hybrid method

hardware steganography

artificial intelligence soware occupationarea energy consumption memory consumption

Figure 13 Prioritizing the implementation techniques of thecryptography algorithms using Q-analysis

Computational Intelligence and Neuroscience 15

Acknowledgments

is work was partially supported by Iran National ScienceFoundation (INSF) under contract No 9653979

References

[1] D Nilesh and M Nagle ldquoe new cryptography algorithmwith high throughputrdquo in Proceedings of the 2014 Interna-tional Conference on Computer Communication and Infor-matics pp 3ndash7 Coimbatore India January 2014

[2] S Ravi P Kocher R Lee G McGraw and A RaghunathanldquoSecurity as a new dimension in embedded system designrdquo inProceedings of the 41st annual conference on Design auto-mation-DAC lsquo04 pp 753ndash760 San Diego CA USA March2004

[3] P Kuppuswamy and S Q Y A Khalidi ldquoHybrid encryptiondecryption technique using new public key and symmetric keyalgorithmrdquo International Journal of Information and Com-puter Security vol 6 no 4 pp 372ndash382 2014

[4] S Bhat and V Kapoor ldquoSecure and efficient data privacyauthentication and integrity schemes using hybrid cryptog-raphyrdquo International Conference on Advanced ComputingNetworking and Informatics vol 870 pp 279ndash285 2019

[5] D Ott and C Peikert ldquoAnd other workshop participantsldquoidentifying research challenges in post quantum cryptog-raphy migration and cryptographic agilityrdquo 2019 httparxivorgabs190907353

[6] P Manickam K Shankar E Perumal M Ilayaraja andK Sathesh Kumar ldquoSecure data transmission through reliablevehicles in VANET using optimal lightweight cryptographyrdquoin Advanced Sciences and Technologies for Security Applica-tions pp 193ndash204 Springer New York NY USA 2019

[7] S Prakash and A Rajput ldquoHybrid cryptography for securedata communication in wireless sensor networksrdquo Advancesin Intelligent Systems and Computing vol 696 pp 589ndash5992018

[8] A A Soofi I Riaz and U Rasheed ldquoAn enhanced vigenerecipher for data securityrdquo International Journal of Scientific ampTechnology Research vol 4 no 8 pp 141ndash145 2015

[9] N Shamsuddin M Syafiqah and S Ali Pitchay ldquoImple-menting location-based cryptography on mobile applicationdesign to secure data in cloud storagerdquo Journal of PhysicsConference Series IOP Publishing vol 1551 no 1 2020

[10] L Jiao H Yonglin and D Feng ldquoStream cipher designs areviewrdquo Science China Information Sciences vol 63 no 3pp 1ndash25 2020

[11] NIST ldquoNIST cryptographic standards and guidelines devel-opment processrdquo Inside NIST vol 27 2016

[12] G Alagic J Alperin-Sheriff D Apon et al Status Report onthe First Round of the NIST Post-Quantum CryptographyStandardization Process National Institute of Standards andTechnology Gaithersburg MD USA pp 1ndash27 2019

[13] I Damaj and S Kasbah ldquoAn analysis framework for hardwareand software implementations with applications from cryp-tographyrdquo Computers amp Electrical Engineering vol 69pp 572ndash584 2018

[14] M S Croock Z A Hassan and S D Khudhur ldquoAdaptive keygeneration algorithm based on software engineering meth-odologyrdquo International Journal of Electrical and ComputerEngineering vol 11 no 1 p 589 2021

[15] K Sha T A Yang W Wei and S Davari ldquoA survey of edgecomputing-based designs for iot securityrdquo Digital Commu-nications and Networks vol 6 no 2 pp 195ndash202 2020

[16] M Yahuza M Y I B Idris A W B A Wahab et alldquoSystematic review on security and privacy requirements inedge computing state of the art and future research oppor-tunitiesrdquo IEEE Access vol 8 pp 76541ndash76567 2020

[17] A Bhowmik J Dey and S Karforma A Novel Double TierCryptographic System (nDTCS) to Reinforce Patientsrsquo Privacyin Contemporary COVID-19 Telemedicine Springer NewYork NY USA 2021

[18] N Gupta User Integrity Protection Security Model forEnhanced Telemedicine for Healthcare Networks 2020httpsosfio69jdq

[19] J Dey A Bhowmik A Sarkar S Karforma andB Chowdhury Cryptographic Engineering on COVID-19Telemedicine An Intelligent Transmission through RecurrentRelation Based Session Key Springer New York NY USA2021

[20] V Hosseinian H Ayatollahi H Haghani and E MehraeenldquoRequirements of information security in a telemedicinenetwork review of IT managersrsquo opinionrdquo Journal of Para-medical Sciences amp Rehabilitation vol 4 no 2 pp 31ndash402015

[21] D Karaoglan Altop A Levi and V Tuzcu ldquoDeriving cryp-tographic keys from physiological signalsrdquo Pervasive andMobile Computing vol 39 pp 65ndash79 2017

[22] O Uzunkol and M S Kiraz ldquoStill wrong use of pairings incryptographyrdquo Applied Mathematics and Computationvol 333 pp 467ndash479 2018

[23] S Picek D Jakobovic J F Miller L Batina and M CupicldquoCryptographic Boolean functions one output many designcriteriardquo Applied Soft Computing vol 40 pp 635ndash653 2016

[24] R Saha and G Geetha ldquoSymmetric random function gen-erator (SRFG) a novel cryptographic primitive for designingfast and robust algorithmsrdquo Chaos Solitons amp Fractalsvol 104 pp 371ndash377 2017

[25] A Achuthshankar and A Achuthshankar ldquoA novel sym-metric cryptography algorithm for fast and secure encryp-tionrdquo in Proceedings of the 2015 IEEE 9th InternationalConference on Intelligent Systems and Control (ISCO) pp 4ndash9Coimbatore India January 2015

[26] D Ott and C Peikert ldquoIdentifying research challenges in postquantum cryptography migration and cryptographic agilityrdquo2019 httparxivorgabs190907353

[27] W Liu B Ying H Yang and H Wang ldquoAccurate modelingfor predicting cryptography overheads on wireless sensornodesrdquovol 2 pp 997ndash1001 in Proceedings of the 2019 11thInternational Conference on Advanced CommunicationTechnology vol 2 pp 997ndash1001 IEEE Bangalore IndiaFebruary 2019

[28] S Feizi A Ahmadi and A Nemati ldquoA hardware imple-mentation of simon cryptography algorithmrdquo in Proceedingsof the 2014 4th International Conference on Computer andKnowledge Engineering (ICCKE) pp 245ndash250 IEEEMashhad Iran 2014

[29] E ambiraja G Ramesh and R Umarani ldquoA survey onvarious most common encryption techniquesrdquo 2012 httpwwwijarcssecom

[30] A Gupta and N K Walia ldquoCryptography Algorithms areviewrdquo International Journal of Engineering Research andDevelopment vol 2 no 2 pp 1667ndash1672 2014

[31] R Bhanot and R Hans ldquoA review and comparative analysis ofvarious encryption algorithmsrdquo International Journal of Se-curity and Its Applications vol 9 no 4 pp 289ndash306 2015

[32] S Chandra S Paira S S Alam and G Sanyal ldquoA com-parative survey of symmetric and asymmetric key

16 Computational Intelligence and Neuroscience

cryptographyrdquo in Proceedings of the 2014 InternationalConference on Electronics Communication and Computa-tional Engineering (ICECCE) pp 83ndash93 Hosur India No-vember 2014

[33] F Maqsood M Ahmed M Mumtaz and M Ali ldquoCryp-tography a comparative analysis for modern techniquesrdquoInternational Journal of Advanced Computer Science andApplications vol 8 no 6 p 6 2017

[34] S B Sadkhan and A O Salman ldquoA survey on lightweight-cryptography status and future challengesrdquo in Proceedings ofthe International Conference on Advances in SustainableEngineering and Applications pp 105ndash108 Wasit Iraq June2018

[35] N V oai ldquoCriteria and dimension reduction of linearmultiple criteria optimization problemsrdquo Journal of GlobalOptimization vol 52 no 3 pp 499ndash508 2012

[36] C E Shannon ldquoCommunication theory of secrecy systems1945rdquo MD Computing vol 15 no 1 pp 57ndash64 1998

[37] L J Fennelly M Beaudry and M A Perry Security in 2025[38] ITU Global Cybersecurity Index 2018 2019[39] ITU Global Cybersecurity Index (GCI) 2017 2017

Computational Intelligence and Neuroscience 17

311 Design of Cryptography Algorithms Using the Cyber-netic Supermodel Considering the above discussion andpresence of numerous indices and components in thedesign and implementation of the cryptography algorithmwhich make it a complex system identifying the rela-tionship between these indices and ranking them is anecessity Accordingly in this section the Q-analysis

method is used and the output of the three matrices isanalyzed According to the level three of the proposedcybernetic model the design and implementation supportprocess of the cryptography algorithms includes 13 com-ponents On the contrary the design and implementationsteps of the cryptography algorithms also include foursteps constituting a 13 lowast 4 matrix

Table 2 e 13-component interaction matrix of the support and the main design and implementation processes of cryptographyalgorithms

Main process Level 1Cryptography algorithms Level 2

Application eoretical basis Implementation Evaluation Level 3

Support process

Software

Culture 5 6 5 2 Organizationstructure 8 7 7 7

International and public relations 5 4 4 7 Financial resources 7 6 7 6

Human resources and education 8 10 10 9 Research and development 10 8 8 10

Rules 7 4 5 10 Management 9 7 8 7

FAVA 10 5 8 7 Standard 10 3 8 10

HardwareEquipment 8 6 7 8

Infrastructure 9 5 5 7Material 7 2 5 5

Table 3 e relationship matrix of the objectives with the design process of the cryptography algorithms

Main process objectives Application eoretical basis Implementation EvaluationSecurity 203 386 516 516Simplicity 5 23 38 38Resources 41 30 81 81Flexibility 15 28 45 45Scalability 6 11 22 22Speed 72 146 247 247Reliability 2 10 11 11

Table 4 A part of the interaction matrix between the cryptography algorithm implementation techniques and the seven objectives

Level 3 Security Simplicity Using resources Flexibility Scalability Speed Reliability

Support process

Avalanche effect 32 0 0 0 0 0 0Digital signature 1 0 0 0 0 0 0

Block size 1 1 0 0 2 0 0Image sharing 0 0 0 1 0 0 0

Parallel processing 0 0 0 1 12 0 0reshold technique 3 1 0 0 0 0 0

Data mining 1 0 0 0 0 0 0Binary tree 1 0 0 0 3 0 0

Cycle 5 2 0 0 12 0 0Multistage crypto 1 0 0 0 0 0 0Hybrid method 30 3 0 2 9 0 2

Hardware 0 0 7 0 0 0 0

6 Computational Intelligence and Neuroscience

32 Calculating the Incidence Matrix First the incidencematric is obtained based on the data matrix 4ndash1 is matrixrepresents the ldquoimpact of support indices on the main designprocess of the cryptography algorithmsrdquo e data matrix iscomprised of two sets D support indices set C and qua-druple design and implementation steps (Table 5) e in-cidence matrix calculated using the data matrix for α 70is represented in Table 6 By assigning different values to theparameter α difference incidence matrices are obtainederesults of the Q-analysis using C++ coding for α 70 aregiven in Figure 4

D d1 d2 d131113864 1113865

C c1 c2 c3 c41113864 1113865(1)

321 Geometric Representation Multidimensional proper-ties of the system are defined by a simple or complex setKD(C λ) such that the entities of the set D represent thesupport indices and entities of the set C represent thequadruple design and implementation steps of the cryp-tography algorithms

In the sample with α70 7 dis is as follows

d1

d2 c1 c2 c3 c41113864 1113865

d3 c41113864 1113865

d4 c1 c31113864 1113865

d5 c1 c2 c3 c41113864 1113865

d6 c1 c2 c3 c41113864 1113865

d7 c1 c41113864 1113865

d8 c1 c2 c3 c41113864 1113865

d9 c1 c3 c41113864 1113865

d10 c1 c3 c41113864 1113865

d11 c1 c3 c41113864 1113865

d12 c1 c41113864 1113865

(2)

e simplexes of σp(di) are also

σminus1 d1( 1113857 σ3 d2( 1113857 σ0 d3( 1113857 σ1 d4( 1113857 σ3 d5( 1113857 σ0 d3( 1113857 σ1 d7( 1113857

σ3 d8( 1113857 σ2 d9( 1113857 σ2 d10( 1113857 σ2 d11( 1113857 σ1 d12( 1113857 σ1 d13( 1113857 (3)

erefore the complex dimension is 3 In other wordsthe diagnosis classes d2 (structureorganization) d5 (humanresourceseducation) d6 (research and development) andd8 (management) have the largest dimension

33 CalculatingDimensions andQ-Link Q-link is defined asthe link between a subset with smallest interface between twosubsequent dis in the chain of d1 to dn Q-link between twosubsequent dis with α70 7 is

Table 5 Sets of dis and cis support indices based on the datamatrix

d1 Culture

d2 Structureorganizationd3 Internationalpublic relationsd4 Financiald5 Human resourceseducationd6 Research and developmentd7 Rulesd8 Managementd9 FAVAd10 Standardd11 Equipmentd12 Infrastructured13 MaterialC1 ApplicationC2 eoretical basisC3 ImplementationC4 Evaluation

Table 6 e incidence matrix of the support indicesrsquo impact of thedesign steps of the cryptography algorithms with α 70

C1 C2 C3 C4

d1 0 0 0 0d2 1 1 1 1d3 0 0 0 1d4 1 0 1 0d5 1 1 1 1d6 1 1 1 1d7 1 0 0 1d8 1 1 1 1d9 1 0 1 1d10 1 0 1 1d11 1 0 1 1d12 1 0 0 1d13 1 0 0 0

Computational Intelligence and Neuroscience 7

σminus1 d1( 1113857 σ3 d2( 1113857⟶ 1 σ3 d2( 1113857 σ0 d3( 1113857⟶ 0 σ0 d3( 1113857 σ1 d4( 1113857⟶ minus 1

σ1 d4( 1113857 σ3 d5( 1113857⟶ 1 σ3 d5( 1113857 σ3 d6( 1113857⟶ 3 σ3 d6( 1113857 σ1 d7( 1113857⟶ 1

σ1 d7( 1113857 σ3 d8( 1113857⟶ 1 σ3 d8( 1113857 σ2 d9( 1113857⟶ 2 σ2 d9( 1113857 σ2 d10( 1113857⟶ 2

σ2 d10( 1113857 σ2 d11( 1113857⟶ 2 σ3 d11( 1113857 σ1 d12( 1113857⟶ 1 σ1 d12( 1113857 σ0 d13( 1113857⟶ 0

(4)

e maximum link dimension is 3 indicating the re-lationship between the diagnosis classes

331 Calculating the Structure Vectors As mentioned inthe definitions the vector Qq is a simplification basiscreated to eliminate the additional impacts in the equiv-alent simplex sets e maximum complex dimension withα70 7 is 3 erefore the first structure vector based onthe output is

Dimension 3 2 1 0 (5)

e second structure vector P is

dimensions 3 2 1 0

P Pdim3 Pdim2 Pdim1 Pdim0( 1113857

P 4 7 10 12( 1113857

(6)

where Pq is the number simplexes greater than or equal to qin the set K in which P is the number of simplex linkrepetitions (support indices) in the quadruple design andimplementation steps of the cryptography algorithms Basedon the values of these two vectors it is seen that the rela-tionship between the support indices and the quadrupledesign and implementation steps of the cryptography al-gorithms is high is issue indicates the role of supportcomponents in the design and implementation of the

cryptography algorithms which should be consideredseriously

332 Calculating the Obstruction or Flexibility VectorQlowastK represents the number of structural obstructions forsimplex interactions in dimension k

Qlowast

Q minus I⟶ Qlowast

1 1 1 11113858 1113859 minus 1 1 1 11113858 1113859⟶ Qlowast

0 0 0 01113858 1113859

(7)

As can be seen there is no obstruction in any of thecommunication levels indicating that there is a significantrelationship between the support components at eachequivalence class

333 Calculating Irregularity e value of (eccprime(σ)) iscalculated using the Chinese method e results for α70

7 are shown in Table 7 e calculated irregularity valueshows that the indices d2 (structureorganization) d5 (hu-man resourceseducation) d6 (research and development)and d8 (management) affect other indices

334 Calculating Complexity Also the results ofQ-analysiscan be used to describe structure complexity According toequations (4)ndash(9) and for α = 7 the complexity measure is

-1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1

3 0 1 3 3 1 3 2 2 2 1

0 -1 0 0 0 0 0 0 0 0

1 1 1 0 1 1 1 1 0

3 3 1 3 2 2 2 1

3 1 3 2 2 2 1

3 2 2 2 1

2 2 2 1

2 2 1

2 1

1

1 1 1 1 1 1

-1

-1

0

0

0

0

0

0

0

0

0

0

0q Q3210

11

11

SETSX2 X5 X6 X8X2 X5 X6 X8 X9 X10 X11X2 X4 X5 X6 X7 X8 X9 X10 X11 X12X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13

Figure 4 Implementation results of the model for support components of the cryptography algorithmsrsquo design with α 70

8 Computational Intelligence and Neuroscience

Q 1 1 1 1( 1113857

ψ(K) 2(1 + 2 + 3 + 4)

(4lowast 5)1113890 1113891 1

(8)

Since in the above model there is no obstruction amongthe component of the equivalent class it was expected thatthe complexity of the support components is not high andthe obtained complexity index of 1 verifies this expectatione system complexity for different alpha-cuts is shown inFigure 5

335 Ranking the Support Components of the Design andImplementation of the Cryptography Algorithms e resultsof using A-analysis are shown in Table 7 e connectionstrength of the factors in one group is specified with alpha-cut erefore the support components are grouped in 5levels Each level describes the priority and importance of thegroup in developing the cryptography algorithms In Fig-ure 6 the ranking pyramid of the support components usingQ-analysis is shown To allocate proper resources thecomponents existing in higher levels of the pyramid (Fig-ure 6) are of higher priority

336 Validation of the Results In this section the resultsof the cybernetic model and Q-analysis for supportcomponentsrsquo ranking are compared with the results re-ported in the global cybersecurity index (GCI) in 20152017 and 2018 presented by ITU [36ndash39] e GCI reportsare focused on five indices including ldquolegal cases orga-nization necessities technical issues capacity buildingand cooperationrdquo and the subindices include legaltechnical organization capacity building and coopera-tion According to the presented indices and subindices itis clear that the ldquorulesrdquo ldquostandardrdquo ldquoresearch and de-velopmentrdquo ldquoeducationrdquo and ldquomanagementrdquo are ofhigher priority in security establishment Although ourresearch is more skilled and detailed compared to the GCIreports but the results verify our findings Also Table 8shows the ranking of the support components obtainedusing Q-analysis

337 Executing the Model and Analyzing the Results of theObjectivesrsquo Impact on Design Steps of the Cryptography Al-gorithmsrsquo Matrix In this section the role of the sevencomponents on the quadruple design and implementationsteps of the cryptography algorithm is analyzed with Q-analysis Using the Q-analysis method and the 7lowast4 matrixobtained from the relationship of the cryptography algo-rithmsrsquo design objectives on their quadruple steps theirindices are ranked

338 Calculating the Incidence Matrix and Executing theModel First the incidence matrix is obtained for the datamatrix shown in Figure 2 It can be seen in Table 9 that eachelement of the two sets represents which indice e inci-dence matrix calculated from the data matrix for α5 40 is

given in Table 10 e results of implementing the model forα5 40 are given in Figure 7

4 Geometric Representation

In the sample with α5 40 dis are

d1 c1 c2 c3 c41113864 1113865

d2

d3 c1 c3 c41113864 1113865

d4 c3 c41113864 1113865

d5

d6 c1 c2 c3 c41113864 1113865

d7

(9)

e simplexes of σp(di) are also

σ3 d1( 1113857σminus1 d2( 1113857σ2 d3( 1113857σ1 d4( 1113857σminus1 d5( 1113857σ3 d6( 1113857σminus1 d7( 1113857

(10)

erefore the complex dimension is 3 In other wordsthe discriminant classes d1 (security) and d6 (speed) have thelargest dimension

41 Calculating the Structure Vectors As mentioned thevector Qq is a simplification basis to eliminate the additionaleffects in the set of equivalent simplexes e maximumcomplex dimension for α5 40 is 3 erefore the firststructure vector based on the output is

Dimension 3 2 1 1

Q 1 1 1 1( 1113857

Dimension 3 2 1 0

P 2 3 4 7( 1113857

(11)

e second structure vector P is

P 2 3 4 7( 1113857 (12)

42 Calculating the Obstruction Vector or InflexibilityQlowastK represents the number of structural obstructions forsimplex interactions in dimension k which is as follows forα5 40

1 1 1 1 1332

005

115

225

com

plex

ity

alpha cutα=5 α=6 α=7 α=8 α=9 α=10

Figure 5 System complexity of the support indices of design andimplementation of the cryptography algorithms for different alpha-cuts

Computational Intelligence and Neuroscience 9

Qlowast

Q minus I⟶ Qlowast

2 1 1 31113858 1113859 minus 1 1 1 11113858 1113859⟶ Qlowast

0 0 0 01113858 1113859

(13)

According to the obtained values it can be concludedthat there are no structural obstructions among the main

design indices of the cryptography algorithms at ismultiple objectives are considered simultaneously by thecryptography algorithm designers

Table 7 e irregularity of the data matrix parameters for α= 7

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d2 d5 d6 d8 qi 3 2 1 364 3 061d9 d10d11 qi 2 1 064 3 011d4 d7 d12 qi 1 014 3 002

human resourceRampD

rules FAVA standard

management infrastructure

organizationstructure equipmentinternational and public relations finanacial

resources and material

Figure 6 e ranking pyramid of the support components using Q-analysis

Table 8 Ranking of the support components using Q-analysis

Relationship of the support components with the design and implementation of the cryptography algorithms (q 0)No relationship α 0 complete relationship α 100

Human resource RampD rules FAVA and standard α100 10Management and infrastructure α90 9Organizationstructure and equipment α80 8International and public relations financial resources and material α70 7Culture α60 6All components α60 5

Table 9 Set of dis and cis the objectivesrsquo indices

eoretical basis C2 Reliability d7 Scalability d5 Resources d3 Security d1Implementation C3 Application C1 Speed d6 Flexibility d4 Simplicity d2Evaluation C4

Table 10 e incidence matrix of the design objectives of thecryptography algorithms with α= 40

c1 c2 c3 c4

d1 1 1 1 1d2 0 0 0 0d3 1 0 1 1d4 0 0 1 1d5 0 0 0 0d6 1 1 1 1d7 0 0 0 0

-1 -1 -1 -1 -1

2 1 -1 2

1 -1 1

-1 -1

3

3 -1 2 1 -1 3 -1

-1

-1

-1

-1

-1

-1q Q3210

11

11

SETSX1X6X1X3X6X1X3X4X6X1X3X4X6

Figure 7 e results of executing the model for design objectivesrsquoindices with α 40

10 Computational Intelligence and Neuroscience

43 Calculating Irregularity e results of applying theChinese method (ecclsquo(σ)) for calculating irregularity forα5 40 are shown in Table 11 e calculated irregu-larity value shows that indices d1 (security) and d6(speed) have received more attention compared to otherindices

44 Calculating Complexity Structural complexity for asample with α5 40 is

dimensions 3 2 1 0

Q Qdim3 Qdim2 Qdim1 Qdim0( 1113857

Q 1 1 1 1( 1113857

ψ(K) 2(1 + 2 + 3 + 4)

(4lowast 5)1113890 1113891 1

(14)

Since in the above model there was no obstructionamong the components of the equivalent classes in any of thecommunication levels it was expected that there is not ahigh complexity among objectives of the cryptography al-gorithm design and the complexity index of 1 verified thisexpectation e system complexity for different alpha-cutsis shown in Figure 8 As can be seen the system complexity is1 for all significant values of alpha

45 Prioritizing the Parameters of the Main Objectives of theCryptography Algorithm Design Using Q-AnalysisAccording to the analysis results the parameters can beclassified into 5 levelsese levels are shown in Table 12 andFigure 9 Each level indicates a priority and importance ofthe group in the development of cryptography algorithmsTo allocate proper resources the components at the higherlevels of the pyramid are of higher priority As can be seenthree objectives of security speed and optimal usage ofresources have the highest priority for the design of cryp-tography algorithms

46 Executing the Model and Analyzing the Results for theCryptography Algorithm Implementation Techniques einteraction matrix between the 29 extracted techniques andthe sevenmain objectives is shown in Figure 10e purposeof this section is to rank the cryptography algorithmimplementation techniques to achieve the goals of interest

47Calculating the IncidenceMatrixandExecuting theModele data matrix A is comprised of two sets e set Drepresents the employed techniques and the set C representsthe seven main design objectives (Table 13)

D d1 d2 d291113864 1113865

C c1 c2 c71113864 1113865(15)

Some parts of the incidence matrix calculated from thedata matrix A for α3 are shown in Table 14 and the resultsof theQ-analysis model for α3 24 are shown in Figure 11

48 Geometric Representation In the sample with α3 24the dis are

d1 c11113864 1113865

d2

d3

d4

d5

d6

d7

d8

d9

d10

d11 c11113864 1113865

d12

d13 c51113864 1113865

d14

d15 c1 c4 c51113864 1113865

d16

d17

d18

d19 c11113864 1113865

d20

d21

d22 c31113864 1113865

d23

d24

d25

d26

d27 c11113864 1113865

d28

d29

(16)

e simplexes of σp(di) are

Computational Intelligence and Neuroscience 11

σ0 d1( 1113857

σminus1 d2( 1113857

σminus1 d3( 1113857

σminus1 d4( 1113857

σminus1 d5( 1113857

σminus1 d6( 1113857

σminus1 d7( 1113857

σminus1 d8( 1113857

σminus1 d9( 1113857

σminus1 d10( 1113857

σ0 d11( 1113857

σminus1 d12( 1113857

σ0 d13( 1113857

σminus1 d14( 1113857

σ2 d15( 1113857

σminus1 d16( 1113857

σminus1 d17( 1113857

σminus1 d18( 1113857

σ0 d19( 1113857

σminus1 d20( 1113857

σminus1 d21( 1113857

σ0 d22( 1113857

σminus1 d23( 1113857

σminus1 d24( 1113857

σminus1 d25( 1113857

σminus1 d26( 1113857

σ0 d27( 1113857

σminus1 d28( 1113857

σminus1 d29( 1113857

(17)

erefore the complex dimension is 2 In other wordsthe discriminant class d15 (basic sciences) has the largestdimension

49 Calculating the Dimensions and the Q-Link In the alphadefined for cryptography algorithm implementation tech-niques to achieve the defined goals the obtained q-linkshows that these techniques are relatively independent al-though there is a weak relationship between some tech-niques Q-link in the samples with α3 24 between eachtwo subsequent di is as follows

σ0 d1( 1113857 σminus1 d2( 1113857⟶ minus 1

σminus1 d2( 1113857 σminus1 d3( 1113857⟶ minus 1

σ0 d3( 1113857 σminus1 d4( 1113857⟶ minus 1

σminus1 d4( 1113857 σminus1 d5( 1113857⟶ minus 1

σminus1 d5( 1113857 σminus1 d6( 1113857⟶ minus 1

σminus1 d6( 1113857 σminus1 d7( 1113857⟶ minus 1

σminus1 d7( 1113857 σminus1 d8( 1113857⟶ minus 1

σminus1 d8( 1113857 σminus1 d9( 1113857⟶ minus 1

σminus1 d9( 1113857 σminus1 d10( 1113857⟶ minus 1

σminus1 d10( 1113857 σ0 d11( 1113857⟶ minus 1

σ0 d11( 1113857 σminus1 d12( 1113857⟶ minus 1

σminus1 d12( 1113857 σ0 d13( 1113857⟶ minus 1

σ0 d13( 1113857 σminus1 d14( 1113857⟶ minus 1

σminus1 d14( 1113857 σ2 d15( 1113857⟶ minus 1

σ2 d15( 1113857 σminus1 d16( 1113857⟶ minus 1

σminus1 d16( 1113857 σminus1 d17( 1113857⟶ minus 1

σminus1 d17( 1113857 σminus1 d18( 1113857⟶ minus 1

σminus1 d18( 1113857 σ0 d19( 1113857⟶ minus 1

σ0 d19( 1113857 σminus1 d20( 1113857⟶ minus 1

σminus1 d20( 1113857 σminus1 d21( 1113857⟶ minus 1

σminus1 d21( 1113857 σ0 d22( 1113857⟶ minus 1

σ0 d22( 1113857 σminus1 d23( 1113857⟶ minus 1

σminus1 d23( 1113857 σminus1 d24( 1113857⟶ minus 1

σminus1 d24( 1113857 σminus1 d25( 1113857⟶ minus 1

σminus1 d25( 1113857 σminus1 d26( 1113857⟶ minus 1

σminus1 d26( 1113857 σ0 d27( 1113857⟶ minus 1

σ0 d27( 1113857 σminus1 d28( 1113857⟶ minus 1

σminus1 d28( 1113857 σminus1 d29( 1113857⟶ minus 1

(18)

Table 11 Irregularity of the cryptography objectivesrsquo parameters in the data matrix A for α5

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d1 and d6 qi 3 2 and 1 433 3 072d3 qi 2 and 1 133 3 026d4 qi 1 033 3 006

11117

1 1 1

09095

1105

11115

12

com

plex

ity

alpha cutα=2 α=3 α=5 α=15 α=25

Figure 8 System complexity of the cryptography algorithm designobjectives for different alpha-cuts

12 Computational Intelligence and Neuroscience

410 Calculating the Structure Vectors erefore the firststructure vector based on the software output is

Q 1 1 2( 1113857 (19)

e second structure vector P is

P 1 1 29( 1113857 (20)

Since the large values of P of higher dimensions dem-onstrate more links the second structure vector calculatedfor the alpha of interest shows that the relationship betweenthe cryptography algorithm implementation techniques isminimum

411 Calculating the Obstruction or Inflexibility Vectore obstruction vector (Qlowast) for a sample with α3 24 is

Table 12 Ranking the objectivesrsquo parameters considering different cuts in Q-analysis

e equivalence class for the parameters with minimum relation level (q 0) αSecurity α50 244Speed α30 244Resources α15 122Flexibility α5 40Simplicity-scalability α2 16(All parameters)

securityspeed

optimal usage of resources

flexibility

simplicity-scalability

Figure 9 Prioritizing the design and implementation objectives of the cryptography algorithms using Q-analysis

securityspeed

optimal usage of resources

flexibility

simplicity-scalability-reliability

Figure 10 Prioritizing the design and implementation objectives of the cryptography algorithms using Q-analysis

Table 13 Set of dis and cis the techniques employed for implementing cryptography algorithms

Flexibility C4 Fuzzy logic d25 Storage space d17 Cycle d9 Avalanche effect d1Scalability C5 Software d26 Clustering d18 Multiple step d10 Digital signature d2Speed C6 Steganography d27 Key d19 Hybrid method d11 Block size d3Reliability C7 Music harmony d28 Graph d20 Hardware d12 Image sharing d4

Artificial intelligence d29 Characteristic oriented d21 Hardware d13 Parallel processing d5Security C1 Energy consumption d22 Occupation area d14 reshold technique d6Simplicity C2 Bandwidth consumption d23 Basic science d15 Data mining d7Resources C3 Memory consumption d24 Compression d16 Binary tree d8

Table 14 e incidence matrix of the cryptography algorithmimplementation techniques with α 24

C1 C2 C3 C4 C5 C6 C7

d1 1 0 0 0 0 0 0d2 0 0 0 0 0 0 0d3 0 0 0 0 0 0 0d4 0 0 0 0 0 0 0d5 0 0 0 0 0 0 0d6 0 0 0 0 0 0 0d7 0 0 0 0 0 0 0d8 0 0 0 0 0 0 0d9 0 0 0 0 0 0 0d10 0 0 0 0 0 0 0d11 1 0 0 0 0 0 0d12 0 0 0 0 0 0 0

Computational Intelligence and Neuroscience 13

Qlowast

Q minus I⟶ Qlowast

1 1 21113858 1113859 minus 1 1 11113858 1113859⟶ Qlowast

0 0 11113858 1113859

(21)

e value of QlowastK represents the number of structurallimitations or obstructions for the cryptography techniquesrsquointeraction at dimension k As can be seen from the cal-culated obstruction vector there is a significant relationshipat some levels

412 Calculating Irregularity Irregularity is the integrationdegree of a cryptography method in the total complex Mea-suring irregularity (eccrsquo) for α3 24 is shown in Table 15 Ascan be seen irregularity for the discriminant class d15 indicatesthat this technique is isolated from other techniques

413 Calculating Complexity Structure complexity for asample with α3 24 is

Q 1 1 2( 1113857

ψ(K) 2(2 + 2 + 3)

(3lowast 4)1113890 1113891 117

(22)

e system complexity for different values of alpha isshown in Figure 12 As can be seen the total complexity ofthe system for large values alpha tends to stability

414 Ranking the Cryptography Algorithm ImplementationTechniques ese techniques can be classified into 5 levelsusing the analysis of the obtained results ese levels can beseen in Table 16 and Figure 13 e link power of the factorsin one group is specified with alpha Each level describespriority and importance of the group in the development ofcryptography algorithms Accordingly six methods of basicscience key management using hardware methods usingsteganography Avalanche effect and using hybrid method

q Q21o

112

SETSX15X15X1X11X15X19X27X13 amp X22

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-10-1-1-10-1-1-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-10-1-1-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

2-10-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1

-1-1-1-1-1-1

-1-1-1-1-1

-1-1-1-1

-1-1-1

-1-10

Figure 11 Results of executing the model for cryptography algorithm implementation techniques with α3 24

14 Computational Intelligence and Neuroscience

are the most important cryptography algorithm imple-mentation techniques to achieve the main objectives

5 Conclusion

In this study the design and implementation model of thecryptography algorithms is designed in three levels and itseffective components are extracted To organize the com-ponents in addition to elimination of the additional com-ponents through the measure reduction algorithm a properclassification is applied to examine the mutual effects Afterclassification three 13lowast 4 7lowast 4 and 29lowast 7 matrices areconstituted and the model is implemented on these threematrices To implement the designed model Q-analysis isused Accordingly for support indices five indices of highpriority include human resources research and develop-ment management organizationstructure and equipmentFor the algorithm design objectives index five high priorityindices include security speed optimal usage of resources

and simplicity For the indices related to implementationtechniques of the cryptography algorithms the most appliedtechniques for achieving the determined objectives includeusing basic science hardware and software methods usingkeymanagement hybrid method steganography Avalancheeffect and using artificial intelligence In the future work wewill consider the following cases

A plan should be formulated according to the prioritiesAccording to the outputs of this study and the presented pri-orities the following topics can be investigated in future studies

(1) Presenting a comprehensive model for generatingvarious cryptography algorithms based on the pri-orities of interest

(2) Examining and formulating a model about the roleof basic science for design and implementation ofcryptography algorithms considering the significantrole of ldquobasic sciencerdquo in implementation of thecryptography algorithms and about the role ofschools and universities

(3) Presenting a model for design and implementationof cryptography algorithms in the IoT infrastructurewith optimal resource usage

Data Availability

e data used to support the findings of the study areavailable from the corresponding author upon request

Conflicts of Interest

e authors declare that there are no conflicts of interestregarding the publication of this article

Table 15 Irregularity of the cryptography algorithm implementation techniques for α3 24

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d15 qi 2 1 3 2 1- qi 1 1 2 033

11117

1 1 1

091

1112

com

plex

ity

alpha-cutα=2 α=3 α=5 α=15 α=25

Figure 12 System complexity of cryptography algorithm implementation techniques for different alpha-cuts

Table 16 Ranking the implementation techniques of the cryptography algorithms considering various alpha-cuts using Q-analysis

Equivalence classes for the parameters with minimum relationship level (q 0) αBasic sciences α20 163Key α10 81Hardware and steganography α5 40Avalanche effect and hybrid method α3 24Artificial intelligence software occupation area energy consumption memory consumption α2 16

basic sciences

key

Avalanche effect hybrid method

hardware steganography

artificial intelligence soware occupationarea energy consumption memory consumption

Figure 13 Prioritizing the implementation techniques of thecryptography algorithms using Q-analysis

Computational Intelligence and Neuroscience 15

Acknowledgments

is work was partially supported by Iran National ScienceFoundation (INSF) under contract No 9653979

References

[1] D Nilesh and M Nagle ldquoe new cryptography algorithmwith high throughputrdquo in Proceedings of the 2014 Interna-tional Conference on Computer Communication and Infor-matics pp 3ndash7 Coimbatore India January 2014

[2] S Ravi P Kocher R Lee G McGraw and A RaghunathanldquoSecurity as a new dimension in embedded system designrdquo inProceedings of the 41st annual conference on Design auto-mation-DAC lsquo04 pp 753ndash760 San Diego CA USA March2004

[3] P Kuppuswamy and S Q Y A Khalidi ldquoHybrid encryptiondecryption technique using new public key and symmetric keyalgorithmrdquo International Journal of Information and Com-puter Security vol 6 no 4 pp 372ndash382 2014

[4] S Bhat and V Kapoor ldquoSecure and efficient data privacyauthentication and integrity schemes using hybrid cryptog-raphyrdquo International Conference on Advanced ComputingNetworking and Informatics vol 870 pp 279ndash285 2019

[5] D Ott and C Peikert ldquoAnd other workshop participantsldquoidentifying research challenges in post quantum cryptog-raphy migration and cryptographic agilityrdquo 2019 httparxivorgabs190907353

[6] P Manickam K Shankar E Perumal M Ilayaraja andK Sathesh Kumar ldquoSecure data transmission through reliablevehicles in VANET using optimal lightweight cryptographyrdquoin Advanced Sciences and Technologies for Security Applica-tions pp 193ndash204 Springer New York NY USA 2019

[7] S Prakash and A Rajput ldquoHybrid cryptography for securedata communication in wireless sensor networksrdquo Advancesin Intelligent Systems and Computing vol 696 pp 589ndash5992018

[8] A A Soofi I Riaz and U Rasheed ldquoAn enhanced vigenerecipher for data securityrdquo International Journal of Scientific ampTechnology Research vol 4 no 8 pp 141ndash145 2015

[9] N Shamsuddin M Syafiqah and S Ali Pitchay ldquoImple-menting location-based cryptography on mobile applicationdesign to secure data in cloud storagerdquo Journal of PhysicsConference Series IOP Publishing vol 1551 no 1 2020

[10] L Jiao H Yonglin and D Feng ldquoStream cipher designs areviewrdquo Science China Information Sciences vol 63 no 3pp 1ndash25 2020

[11] NIST ldquoNIST cryptographic standards and guidelines devel-opment processrdquo Inside NIST vol 27 2016

[12] G Alagic J Alperin-Sheriff D Apon et al Status Report onthe First Round of the NIST Post-Quantum CryptographyStandardization Process National Institute of Standards andTechnology Gaithersburg MD USA pp 1ndash27 2019

[13] I Damaj and S Kasbah ldquoAn analysis framework for hardwareand software implementations with applications from cryp-tographyrdquo Computers amp Electrical Engineering vol 69pp 572ndash584 2018

[14] M S Croock Z A Hassan and S D Khudhur ldquoAdaptive keygeneration algorithm based on software engineering meth-odologyrdquo International Journal of Electrical and ComputerEngineering vol 11 no 1 p 589 2021

[15] K Sha T A Yang W Wei and S Davari ldquoA survey of edgecomputing-based designs for iot securityrdquo Digital Commu-nications and Networks vol 6 no 2 pp 195ndash202 2020

[16] M Yahuza M Y I B Idris A W B A Wahab et alldquoSystematic review on security and privacy requirements inedge computing state of the art and future research oppor-tunitiesrdquo IEEE Access vol 8 pp 76541ndash76567 2020

[17] A Bhowmik J Dey and S Karforma A Novel Double TierCryptographic System (nDTCS) to Reinforce Patientsrsquo Privacyin Contemporary COVID-19 Telemedicine Springer NewYork NY USA 2021

[18] N Gupta User Integrity Protection Security Model forEnhanced Telemedicine for Healthcare Networks 2020httpsosfio69jdq

[19] J Dey A Bhowmik A Sarkar S Karforma andB Chowdhury Cryptographic Engineering on COVID-19Telemedicine An Intelligent Transmission through RecurrentRelation Based Session Key Springer New York NY USA2021

[20] V Hosseinian H Ayatollahi H Haghani and E MehraeenldquoRequirements of information security in a telemedicinenetwork review of IT managersrsquo opinionrdquo Journal of Para-medical Sciences amp Rehabilitation vol 4 no 2 pp 31ndash402015

[21] D Karaoglan Altop A Levi and V Tuzcu ldquoDeriving cryp-tographic keys from physiological signalsrdquo Pervasive andMobile Computing vol 39 pp 65ndash79 2017

[22] O Uzunkol and M S Kiraz ldquoStill wrong use of pairings incryptographyrdquo Applied Mathematics and Computationvol 333 pp 467ndash479 2018

[23] S Picek D Jakobovic J F Miller L Batina and M CupicldquoCryptographic Boolean functions one output many designcriteriardquo Applied Soft Computing vol 40 pp 635ndash653 2016

[24] R Saha and G Geetha ldquoSymmetric random function gen-erator (SRFG) a novel cryptographic primitive for designingfast and robust algorithmsrdquo Chaos Solitons amp Fractalsvol 104 pp 371ndash377 2017

[25] A Achuthshankar and A Achuthshankar ldquoA novel sym-metric cryptography algorithm for fast and secure encryp-tionrdquo in Proceedings of the 2015 IEEE 9th InternationalConference on Intelligent Systems and Control (ISCO) pp 4ndash9Coimbatore India January 2015

[26] D Ott and C Peikert ldquoIdentifying research challenges in postquantum cryptography migration and cryptographic agilityrdquo2019 httparxivorgabs190907353

[27] W Liu B Ying H Yang and H Wang ldquoAccurate modelingfor predicting cryptography overheads on wireless sensornodesrdquovol 2 pp 997ndash1001 in Proceedings of the 2019 11thInternational Conference on Advanced CommunicationTechnology vol 2 pp 997ndash1001 IEEE Bangalore IndiaFebruary 2019

[28] S Feizi A Ahmadi and A Nemati ldquoA hardware imple-mentation of simon cryptography algorithmrdquo in Proceedingsof the 2014 4th International Conference on Computer andKnowledge Engineering (ICCKE) pp 245ndash250 IEEEMashhad Iran 2014

[29] E ambiraja G Ramesh and R Umarani ldquoA survey onvarious most common encryption techniquesrdquo 2012 httpwwwijarcssecom

[30] A Gupta and N K Walia ldquoCryptography Algorithms areviewrdquo International Journal of Engineering Research andDevelopment vol 2 no 2 pp 1667ndash1672 2014

[31] R Bhanot and R Hans ldquoA review and comparative analysis ofvarious encryption algorithmsrdquo International Journal of Se-curity and Its Applications vol 9 no 4 pp 289ndash306 2015

[32] S Chandra S Paira S S Alam and G Sanyal ldquoA com-parative survey of symmetric and asymmetric key

16 Computational Intelligence and Neuroscience

cryptographyrdquo in Proceedings of the 2014 InternationalConference on Electronics Communication and Computa-tional Engineering (ICECCE) pp 83ndash93 Hosur India No-vember 2014

[33] F Maqsood M Ahmed M Mumtaz and M Ali ldquoCryp-tography a comparative analysis for modern techniquesrdquoInternational Journal of Advanced Computer Science andApplications vol 8 no 6 p 6 2017

[34] S B Sadkhan and A O Salman ldquoA survey on lightweight-cryptography status and future challengesrdquo in Proceedings ofthe International Conference on Advances in SustainableEngineering and Applications pp 105ndash108 Wasit Iraq June2018

[35] N V oai ldquoCriteria and dimension reduction of linearmultiple criteria optimization problemsrdquo Journal of GlobalOptimization vol 52 no 3 pp 499ndash508 2012

[36] C E Shannon ldquoCommunication theory of secrecy systems1945rdquo MD Computing vol 15 no 1 pp 57ndash64 1998

[37] L J Fennelly M Beaudry and M A Perry Security in 2025[38] ITU Global Cybersecurity Index 2018 2019[39] ITU Global Cybersecurity Index (GCI) 2017 2017

Computational Intelligence and Neuroscience 17

32 Calculating the Incidence Matrix First the incidencematric is obtained based on the data matrix 4ndash1 is matrixrepresents the ldquoimpact of support indices on the main designprocess of the cryptography algorithmsrdquo e data matrix iscomprised of two sets D support indices set C and qua-druple design and implementation steps (Table 5) e in-cidence matrix calculated using the data matrix for α 70is represented in Table 6 By assigning different values to theparameter α difference incidence matrices are obtainederesults of the Q-analysis using C++ coding for α 70 aregiven in Figure 4

D d1 d2 d131113864 1113865

C c1 c2 c3 c41113864 1113865(1)

321 Geometric Representation Multidimensional proper-ties of the system are defined by a simple or complex setKD(C λ) such that the entities of the set D represent thesupport indices and entities of the set C represent thequadruple design and implementation steps of the cryp-tography algorithms

In the sample with α70 7 dis is as follows

d1

d2 c1 c2 c3 c41113864 1113865

d3 c41113864 1113865

d4 c1 c31113864 1113865

d5 c1 c2 c3 c41113864 1113865

d6 c1 c2 c3 c41113864 1113865

d7 c1 c41113864 1113865

d8 c1 c2 c3 c41113864 1113865

d9 c1 c3 c41113864 1113865

d10 c1 c3 c41113864 1113865

d11 c1 c3 c41113864 1113865

d12 c1 c41113864 1113865

(2)

e simplexes of σp(di) are also

σminus1 d1( 1113857 σ3 d2( 1113857 σ0 d3( 1113857 σ1 d4( 1113857 σ3 d5( 1113857 σ0 d3( 1113857 σ1 d7( 1113857

σ3 d8( 1113857 σ2 d9( 1113857 σ2 d10( 1113857 σ2 d11( 1113857 σ1 d12( 1113857 σ1 d13( 1113857 (3)

erefore the complex dimension is 3 In other wordsthe diagnosis classes d2 (structureorganization) d5 (humanresourceseducation) d6 (research and development) andd8 (management) have the largest dimension

33 CalculatingDimensions andQ-Link Q-link is defined asthe link between a subset with smallest interface between twosubsequent dis in the chain of d1 to dn Q-link between twosubsequent dis with α70 7 is

Table 5 Sets of dis and cis support indices based on the datamatrix

d1 Culture

d2 Structureorganizationd3 Internationalpublic relationsd4 Financiald5 Human resourceseducationd6 Research and developmentd7 Rulesd8 Managementd9 FAVAd10 Standardd11 Equipmentd12 Infrastructured13 MaterialC1 ApplicationC2 eoretical basisC3 ImplementationC4 Evaluation

Table 6 e incidence matrix of the support indicesrsquo impact of thedesign steps of the cryptography algorithms with α 70

C1 C2 C3 C4

d1 0 0 0 0d2 1 1 1 1d3 0 0 0 1d4 1 0 1 0d5 1 1 1 1d6 1 1 1 1d7 1 0 0 1d8 1 1 1 1d9 1 0 1 1d10 1 0 1 1d11 1 0 1 1d12 1 0 0 1d13 1 0 0 0

Computational Intelligence and Neuroscience 7

σminus1 d1( 1113857 σ3 d2( 1113857⟶ 1 σ3 d2( 1113857 σ0 d3( 1113857⟶ 0 σ0 d3( 1113857 σ1 d4( 1113857⟶ minus 1

σ1 d4( 1113857 σ3 d5( 1113857⟶ 1 σ3 d5( 1113857 σ3 d6( 1113857⟶ 3 σ3 d6( 1113857 σ1 d7( 1113857⟶ 1

σ1 d7( 1113857 σ3 d8( 1113857⟶ 1 σ3 d8( 1113857 σ2 d9( 1113857⟶ 2 σ2 d9( 1113857 σ2 d10( 1113857⟶ 2

σ2 d10( 1113857 σ2 d11( 1113857⟶ 2 σ3 d11( 1113857 σ1 d12( 1113857⟶ 1 σ1 d12( 1113857 σ0 d13( 1113857⟶ 0

(4)

e maximum link dimension is 3 indicating the re-lationship between the diagnosis classes

331 Calculating the Structure Vectors As mentioned inthe definitions the vector Qq is a simplification basiscreated to eliminate the additional impacts in the equiv-alent simplex sets e maximum complex dimension withα70 7 is 3 erefore the first structure vector based onthe output is

Dimension 3 2 1 0 (5)

e second structure vector P is

dimensions 3 2 1 0

P Pdim3 Pdim2 Pdim1 Pdim0( 1113857

P 4 7 10 12( 1113857

(6)

where Pq is the number simplexes greater than or equal to qin the set K in which P is the number of simplex linkrepetitions (support indices) in the quadruple design andimplementation steps of the cryptography algorithms Basedon the values of these two vectors it is seen that the rela-tionship between the support indices and the quadrupledesign and implementation steps of the cryptography al-gorithms is high is issue indicates the role of supportcomponents in the design and implementation of the

cryptography algorithms which should be consideredseriously

332 Calculating the Obstruction or Flexibility VectorQlowastK represents the number of structural obstructions forsimplex interactions in dimension k

Qlowast

Q minus I⟶ Qlowast

1 1 1 11113858 1113859 minus 1 1 1 11113858 1113859⟶ Qlowast

0 0 0 01113858 1113859

(7)

As can be seen there is no obstruction in any of thecommunication levels indicating that there is a significantrelationship between the support components at eachequivalence class

333 Calculating Irregularity e value of (eccprime(σ)) iscalculated using the Chinese method e results for α70

7 are shown in Table 7 e calculated irregularity valueshows that the indices d2 (structureorganization) d5 (hu-man resourceseducation) d6 (research and development)and d8 (management) affect other indices

334 Calculating Complexity Also the results ofQ-analysiscan be used to describe structure complexity According toequations (4)ndash(9) and for α = 7 the complexity measure is

-1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1

3 0 1 3 3 1 3 2 2 2 1

0 -1 0 0 0 0 0 0 0 0

1 1 1 0 1 1 1 1 0

3 3 1 3 2 2 2 1

3 1 3 2 2 2 1

3 2 2 2 1

2 2 2 1

2 2 1

2 1

1

1 1 1 1 1 1

-1

-1

0

0

0

0

0

0

0

0

0

0

0q Q3210

11

11

SETSX2 X5 X6 X8X2 X5 X6 X8 X9 X10 X11X2 X4 X5 X6 X7 X8 X9 X10 X11 X12X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13

Figure 4 Implementation results of the model for support components of the cryptography algorithmsrsquo design with α 70

8 Computational Intelligence and Neuroscience

Q 1 1 1 1( 1113857

ψ(K) 2(1 + 2 + 3 + 4)

(4lowast 5)1113890 1113891 1

(8)

Since in the above model there is no obstruction amongthe component of the equivalent class it was expected thatthe complexity of the support components is not high andthe obtained complexity index of 1 verifies this expectatione system complexity for different alpha-cuts is shown inFigure 5

335 Ranking the Support Components of the Design andImplementation of the Cryptography Algorithms e resultsof using A-analysis are shown in Table 7 e connectionstrength of the factors in one group is specified with alpha-cut erefore the support components are grouped in 5levels Each level describes the priority and importance of thegroup in developing the cryptography algorithms In Fig-ure 6 the ranking pyramid of the support components usingQ-analysis is shown To allocate proper resources thecomponents existing in higher levels of the pyramid (Fig-ure 6) are of higher priority

336 Validation of the Results In this section the resultsof the cybernetic model and Q-analysis for supportcomponentsrsquo ranking are compared with the results re-ported in the global cybersecurity index (GCI) in 20152017 and 2018 presented by ITU [36ndash39] e GCI reportsare focused on five indices including ldquolegal cases orga-nization necessities technical issues capacity buildingand cooperationrdquo and the subindices include legaltechnical organization capacity building and coopera-tion According to the presented indices and subindices itis clear that the ldquorulesrdquo ldquostandardrdquo ldquoresearch and de-velopmentrdquo ldquoeducationrdquo and ldquomanagementrdquo are ofhigher priority in security establishment Although ourresearch is more skilled and detailed compared to the GCIreports but the results verify our findings Also Table 8shows the ranking of the support components obtainedusing Q-analysis

337 Executing the Model and Analyzing the Results of theObjectivesrsquo Impact on Design Steps of the Cryptography Al-gorithmsrsquo Matrix In this section the role of the sevencomponents on the quadruple design and implementationsteps of the cryptography algorithm is analyzed with Q-analysis Using the Q-analysis method and the 7lowast4 matrixobtained from the relationship of the cryptography algo-rithmsrsquo design objectives on their quadruple steps theirindices are ranked

338 Calculating the Incidence Matrix and Executing theModel First the incidence matrix is obtained for the datamatrix shown in Figure 2 It can be seen in Table 9 that eachelement of the two sets represents which indice e inci-dence matrix calculated from the data matrix for α5 40 is

given in Table 10 e results of implementing the model forα5 40 are given in Figure 7

4 Geometric Representation

In the sample with α5 40 dis are

d1 c1 c2 c3 c41113864 1113865

d2

d3 c1 c3 c41113864 1113865

d4 c3 c41113864 1113865

d5

d6 c1 c2 c3 c41113864 1113865

d7

(9)

e simplexes of σp(di) are also

σ3 d1( 1113857σminus1 d2( 1113857σ2 d3( 1113857σ1 d4( 1113857σminus1 d5( 1113857σ3 d6( 1113857σminus1 d7( 1113857

(10)

erefore the complex dimension is 3 In other wordsthe discriminant classes d1 (security) and d6 (speed) have thelargest dimension

41 Calculating the Structure Vectors As mentioned thevector Qq is a simplification basis to eliminate the additionaleffects in the set of equivalent simplexes e maximumcomplex dimension for α5 40 is 3 erefore the firststructure vector based on the output is

Dimension 3 2 1 1

Q 1 1 1 1( 1113857

Dimension 3 2 1 0

P 2 3 4 7( 1113857

(11)

e second structure vector P is

P 2 3 4 7( 1113857 (12)

42 Calculating the Obstruction Vector or InflexibilityQlowastK represents the number of structural obstructions forsimplex interactions in dimension k which is as follows forα5 40

1 1 1 1 1332

005

115

225

com

plex

ity

alpha cutα=5 α=6 α=7 α=8 α=9 α=10

Figure 5 System complexity of the support indices of design andimplementation of the cryptography algorithms for different alpha-cuts

Computational Intelligence and Neuroscience 9

Qlowast

Q minus I⟶ Qlowast

2 1 1 31113858 1113859 minus 1 1 1 11113858 1113859⟶ Qlowast

0 0 0 01113858 1113859

(13)

According to the obtained values it can be concludedthat there are no structural obstructions among the main

design indices of the cryptography algorithms at ismultiple objectives are considered simultaneously by thecryptography algorithm designers

Table 7 e irregularity of the data matrix parameters for α= 7

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d2 d5 d6 d8 qi 3 2 1 364 3 061d9 d10d11 qi 2 1 064 3 011d4 d7 d12 qi 1 014 3 002

human resourceRampD

rules FAVA standard

management infrastructure

organizationstructure equipmentinternational and public relations finanacial

resources and material

Figure 6 e ranking pyramid of the support components using Q-analysis

Table 8 Ranking of the support components using Q-analysis

Relationship of the support components with the design and implementation of the cryptography algorithms (q 0)No relationship α 0 complete relationship α 100

Human resource RampD rules FAVA and standard α100 10Management and infrastructure α90 9Organizationstructure and equipment α80 8International and public relations financial resources and material α70 7Culture α60 6All components α60 5

Table 9 Set of dis and cis the objectivesrsquo indices

eoretical basis C2 Reliability d7 Scalability d5 Resources d3 Security d1Implementation C3 Application C1 Speed d6 Flexibility d4 Simplicity d2Evaluation C4

Table 10 e incidence matrix of the design objectives of thecryptography algorithms with α= 40

c1 c2 c3 c4

d1 1 1 1 1d2 0 0 0 0d3 1 0 1 1d4 0 0 1 1d5 0 0 0 0d6 1 1 1 1d7 0 0 0 0

-1 -1 -1 -1 -1

2 1 -1 2

1 -1 1

-1 -1

3

3 -1 2 1 -1 3 -1

-1

-1

-1

-1

-1

-1q Q3210

11

11

SETSX1X6X1X3X6X1X3X4X6X1X3X4X6

Figure 7 e results of executing the model for design objectivesrsquoindices with α 40

10 Computational Intelligence and Neuroscience

43 Calculating Irregularity e results of applying theChinese method (ecclsquo(σ)) for calculating irregularity forα5 40 are shown in Table 11 e calculated irregu-larity value shows that indices d1 (security) and d6(speed) have received more attention compared to otherindices

44 Calculating Complexity Structural complexity for asample with α5 40 is

dimensions 3 2 1 0

Q Qdim3 Qdim2 Qdim1 Qdim0( 1113857

Q 1 1 1 1( 1113857

ψ(K) 2(1 + 2 + 3 + 4)

(4lowast 5)1113890 1113891 1

(14)

Since in the above model there was no obstructionamong the components of the equivalent classes in any of thecommunication levels it was expected that there is not ahigh complexity among objectives of the cryptography al-gorithm design and the complexity index of 1 verified thisexpectation e system complexity for different alpha-cutsis shown in Figure 8 As can be seen the system complexity is1 for all significant values of alpha

45 Prioritizing the Parameters of the Main Objectives of theCryptography Algorithm Design Using Q-AnalysisAccording to the analysis results the parameters can beclassified into 5 levelsese levels are shown in Table 12 andFigure 9 Each level indicates a priority and importance ofthe group in the development of cryptography algorithmsTo allocate proper resources the components at the higherlevels of the pyramid are of higher priority As can be seenthree objectives of security speed and optimal usage ofresources have the highest priority for the design of cryp-tography algorithms

46 Executing the Model and Analyzing the Results for theCryptography Algorithm Implementation Techniques einteraction matrix between the 29 extracted techniques andthe sevenmain objectives is shown in Figure 10e purposeof this section is to rank the cryptography algorithmimplementation techniques to achieve the goals of interest

47Calculating the IncidenceMatrixandExecuting theModele data matrix A is comprised of two sets e set Drepresents the employed techniques and the set C representsthe seven main design objectives (Table 13)

D d1 d2 d291113864 1113865

C c1 c2 c71113864 1113865(15)

Some parts of the incidence matrix calculated from thedata matrix A for α3 are shown in Table 14 and the resultsof theQ-analysis model for α3 24 are shown in Figure 11

48 Geometric Representation In the sample with α3 24the dis are

d1 c11113864 1113865

d2

d3

d4

d5

d6

d7

d8

d9

d10

d11 c11113864 1113865

d12

d13 c51113864 1113865

d14

d15 c1 c4 c51113864 1113865

d16

d17

d18

d19 c11113864 1113865

d20

d21

d22 c31113864 1113865

d23

d24

d25

d26

d27 c11113864 1113865

d28

d29

(16)

e simplexes of σp(di) are

Computational Intelligence and Neuroscience 11

σ0 d1( 1113857

σminus1 d2( 1113857

σminus1 d3( 1113857

σminus1 d4( 1113857

σminus1 d5( 1113857

σminus1 d6( 1113857

σminus1 d7( 1113857

σminus1 d8( 1113857

σminus1 d9( 1113857

σminus1 d10( 1113857

σ0 d11( 1113857

σminus1 d12( 1113857

σ0 d13( 1113857

σminus1 d14( 1113857

σ2 d15( 1113857

σminus1 d16( 1113857

σminus1 d17( 1113857

σminus1 d18( 1113857

σ0 d19( 1113857

σminus1 d20( 1113857

σminus1 d21( 1113857

σ0 d22( 1113857

σminus1 d23( 1113857

σminus1 d24( 1113857

σminus1 d25( 1113857

σminus1 d26( 1113857

σ0 d27( 1113857

σminus1 d28( 1113857

σminus1 d29( 1113857

(17)

erefore the complex dimension is 2 In other wordsthe discriminant class d15 (basic sciences) has the largestdimension

49 Calculating the Dimensions and the Q-Link In the alphadefined for cryptography algorithm implementation tech-niques to achieve the defined goals the obtained q-linkshows that these techniques are relatively independent al-though there is a weak relationship between some tech-niques Q-link in the samples with α3 24 between eachtwo subsequent di is as follows

σ0 d1( 1113857 σminus1 d2( 1113857⟶ minus 1

σminus1 d2( 1113857 σminus1 d3( 1113857⟶ minus 1

σ0 d3( 1113857 σminus1 d4( 1113857⟶ minus 1

σminus1 d4( 1113857 σminus1 d5( 1113857⟶ minus 1

σminus1 d5( 1113857 σminus1 d6( 1113857⟶ minus 1

σminus1 d6( 1113857 σminus1 d7( 1113857⟶ minus 1

σminus1 d7( 1113857 σminus1 d8( 1113857⟶ minus 1

σminus1 d8( 1113857 σminus1 d9( 1113857⟶ minus 1

σminus1 d9( 1113857 σminus1 d10( 1113857⟶ minus 1

σminus1 d10( 1113857 σ0 d11( 1113857⟶ minus 1

σ0 d11( 1113857 σminus1 d12( 1113857⟶ minus 1

σminus1 d12( 1113857 σ0 d13( 1113857⟶ minus 1

σ0 d13( 1113857 σminus1 d14( 1113857⟶ minus 1

σminus1 d14( 1113857 σ2 d15( 1113857⟶ minus 1

σ2 d15( 1113857 σminus1 d16( 1113857⟶ minus 1

σminus1 d16( 1113857 σminus1 d17( 1113857⟶ minus 1

σminus1 d17( 1113857 σminus1 d18( 1113857⟶ minus 1

σminus1 d18( 1113857 σ0 d19( 1113857⟶ minus 1

σ0 d19( 1113857 σminus1 d20( 1113857⟶ minus 1

σminus1 d20( 1113857 σminus1 d21( 1113857⟶ minus 1

σminus1 d21( 1113857 σ0 d22( 1113857⟶ minus 1

σ0 d22( 1113857 σminus1 d23( 1113857⟶ minus 1

σminus1 d23( 1113857 σminus1 d24( 1113857⟶ minus 1

σminus1 d24( 1113857 σminus1 d25( 1113857⟶ minus 1

σminus1 d25( 1113857 σminus1 d26( 1113857⟶ minus 1

σminus1 d26( 1113857 σ0 d27( 1113857⟶ minus 1

σ0 d27( 1113857 σminus1 d28( 1113857⟶ minus 1

σminus1 d28( 1113857 σminus1 d29( 1113857⟶ minus 1

(18)

Table 11 Irregularity of the cryptography objectivesrsquo parameters in the data matrix A for α5

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d1 and d6 qi 3 2 and 1 433 3 072d3 qi 2 and 1 133 3 026d4 qi 1 033 3 006

11117

1 1 1

09095

1105

11115

12

com

plex

ity

alpha cutα=2 α=3 α=5 α=15 α=25

Figure 8 System complexity of the cryptography algorithm designobjectives for different alpha-cuts

12 Computational Intelligence and Neuroscience

410 Calculating the Structure Vectors erefore the firststructure vector based on the software output is

Q 1 1 2( 1113857 (19)

e second structure vector P is

P 1 1 29( 1113857 (20)

Since the large values of P of higher dimensions dem-onstrate more links the second structure vector calculatedfor the alpha of interest shows that the relationship betweenthe cryptography algorithm implementation techniques isminimum

411 Calculating the Obstruction or Inflexibility Vectore obstruction vector (Qlowast) for a sample with α3 24 is

Table 12 Ranking the objectivesrsquo parameters considering different cuts in Q-analysis

e equivalence class for the parameters with minimum relation level (q 0) αSecurity α50 244Speed α30 244Resources α15 122Flexibility α5 40Simplicity-scalability α2 16(All parameters)

securityspeed

optimal usage of resources

flexibility

simplicity-scalability

Figure 9 Prioritizing the design and implementation objectives of the cryptography algorithms using Q-analysis

securityspeed

optimal usage of resources

flexibility

simplicity-scalability-reliability

Figure 10 Prioritizing the design and implementation objectives of the cryptography algorithms using Q-analysis

Table 13 Set of dis and cis the techniques employed for implementing cryptography algorithms

Flexibility C4 Fuzzy logic d25 Storage space d17 Cycle d9 Avalanche effect d1Scalability C5 Software d26 Clustering d18 Multiple step d10 Digital signature d2Speed C6 Steganography d27 Key d19 Hybrid method d11 Block size d3Reliability C7 Music harmony d28 Graph d20 Hardware d12 Image sharing d4

Artificial intelligence d29 Characteristic oriented d21 Hardware d13 Parallel processing d5Security C1 Energy consumption d22 Occupation area d14 reshold technique d6Simplicity C2 Bandwidth consumption d23 Basic science d15 Data mining d7Resources C3 Memory consumption d24 Compression d16 Binary tree d8

Table 14 e incidence matrix of the cryptography algorithmimplementation techniques with α 24

C1 C2 C3 C4 C5 C6 C7

d1 1 0 0 0 0 0 0d2 0 0 0 0 0 0 0d3 0 0 0 0 0 0 0d4 0 0 0 0 0 0 0d5 0 0 0 0 0 0 0d6 0 0 0 0 0 0 0d7 0 0 0 0 0 0 0d8 0 0 0 0 0 0 0d9 0 0 0 0 0 0 0d10 0 0 0 0 0 0 0d11 1 0 0 0 0 0 0d12 0 0 0 0 0 0 0

Computational Intelligence and Neuroscience 13

Qlowast

Q minus I⟶ Qlowast

1 1 21113858 1113859 minus 1 1 11113858 1113859⟶ Qlowast

0 0 11113858 1113859

(21)

e value of QlowastK represents the number of structurallimitations or obstructions for the cryptography techniquesrsquointeraction at dimension k As can be seen from the cal-culated obstruction vector there is a significant relationshipat some levels

412 Calculating Irregularity Irregularity is the integrationdegree of a cryptography method in the total complex Mea-suring irregularity (eccrsquo) for α3 24 is shown in Table 15 Ascan be seen irregularity for the discriminant class d15 indicatesthat this technique is isolated from other techniques

413 Calculating Complexity Structure complexity for asample with α3 24 is

Q 1 1 2( 1113857

ψ(K) 2(2 + 2 + 3)

(3lowast 4)1113890 1113891 117

(22)

e system complexity for different values of alpha isshown in Figure 12 As can be seen the total complexity ofthe system for large values alpha tends to stability

414 Ranking the Cryptography Algorithm ImplementationTechniques ese techniques can be classified into 5 levelsusing the analysis of the obtained results ese levels can beseen in Table 16 and Figure 13 e link power of the factorsin one group is specified with alpha Each level describespriority and importance of the group in the development ofcryptography algorithms Accordingly six methods of basicscience key management using hardware methods usingsteganography Avalanche effect and using hybrid method

q Q21o

112

SETSX15X15X1X11X15X19X27X13 amp X22

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-10-1-1-10-1-1-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-10-1-1-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

2-10-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1

-1-1-1-1-1-1

-1-1-1-1-1

-1-1-1-1

-1-1-1

-1-10

Figure 11 Results of executing the model for cryptography algorithm implementation techniques with α3 24

14 Computational Intelligence and Neuroscience

are the most important cryptography algorithm imple-mentation techniques to achieve the main objectives

5 Conclusion

In this study the design and implementation model of thecryptography algorithms is designed in three levels and itseffective components are extracted To organize the com-ponents in addition to elimination of the additional com-ponents through the measure reduction algorithm a properclassification is applied to examine the mutual effects Afterclassification three 13lowast 4 7lowast 4 and 29lowast 7 matrices areconstituted and the model is implemented on these threematrices To implement the designed model Q-analysis isused Accordingly for support indices five indices of highpriority include human resources research and develop-ment management organizationstructure and equipmentFor the algorithm design objectives index five high priorityindices include security speed optimal usage of resources

and simplicity For the indices related to implementationtechniques of the cryptography algorithms the most appliedtechniques for achieving the determined objectives includeusing basic science hardware and software methods usingkeymanagement hybrid method steganography Avalancheeffect and using artificial intelligence In the future work wewill consider the following cases

A plan should be formulated according to the prioritiesAccording to the outputs of this study and the presented pri-orities the following topics can be investigated in future studies

(1) Presenting a comprehensive model for generatingvarious cryptography algorithms based on the pri-orities of interest

(2) Examining and formulating a model about the roleof basic science for design and implementation ofcryptography algorithms considering the significantrole of ldquobasic sciencerdquo in implementation of thecryptography algorithms and about the role ofschools and universities

(3) Presenting a model for design and implementationof cryptography algorithms in the IoT infrastructurewith optimal resource usage

Data Availability

e data used to support the findings of the study areavailable from the corresponding author upon request

Conflicts of Interest

e authors declare that there are no conflicts of interestregarding the publication of this article

Table 15 Irregularity of the cryptography algorithm implementation techniques for α3 24

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d15 qi 2 1 3 2 1- qi 1 1 2 033

11117

1 1 1

091

1112

com

plex

ity

alpha-cutα=2 α=3 α=5 α=15 α=25

Figure 12 System complexity of cryptography algorithm implementation techniques for different alpha-cuts

Table 16 Ranking the implementation techniques of the cryptography algorithms considering various alpha-cuts using Q-analysis

Equivalence classes for the parameters with minimum relationship level (q 0) αBasic sciences α20 163Key α10 81Hardware and steganography α5 40Avalanche effect and hybrid method α3 24Artificial intelligence software occupation area energy consumption memory consumption α2 16

basic sciences

key

Avalanche effect hybrid method

hardware steganography

artificial intelligence soware occupationarea energy consumption memory consumption

Figure 13 Prioritizing the implementation techniques of thecryptography algorithms using Q-analysis

Computational Intelligence and Neuroscience 15

Acknowledgments

is work was partially supported by Iran National ScienceFoundation (INSF) under contract No 9653979

References

[1] D Nilesh and M Nagle ldquoe new cryptography algorithmwith high throughputrdquo in Proceedings of the 2014 Interna-tional Conference on Computer Communication and Infor-matics pp 3ndash7 Coimbatore India January 2014

[2] S Ravi P Kocher R Lee G McGraw and A RaghunathanldquoSecurity as a new dimension in embedded system designrdquo inProceedings of the 41st annual conference on Design auto-mation-DAC lsquo04 pp 753ndash760 San Diego CA USA March2004

[3] P Kuppuswamy and S Q Y A Khalidi ldquoHybrid encryptiondecryption technique using new public key and symmetric keyalgorithmrdquo International Journal of Information and Com-puter Security vol 6 no 4 pp 372ndash382 2014

[4] S Bhat and V Kapoor ldquoSecure and efficient data privacyauthentication and integrity schemes using hybrid cryptog-raphyrdquo International Conference on Advanced ComputingNetworking and Informatics vol 870 pp 279ndash285 2019

[5] D Ott and C Peikert ldquoAnd other workshop participantsldquoidentifying research challenges in post quantum cryptog-raphy migration and cryptographic agilityrdquo 2019 httparxivorgabs190907353

[6] P Manickam K Shankar E Perumal M Ilayaraja andK Sathesh Kumar ldquoSecure data transmission through reliablevehicles in VANET using optimal lightweight cryptographyrdquoin Advanced Sciences and Technologies for Security Applica-tions pp 193ndash204 Springer New York NY USA 2019

[7] S Prakash and A Rajput ldquoHybrid cryptography for securedata communication in wireless sensor networksrdquo Advancesin Intelligent Systems and Computing vol 696 pp 589ndash5992018

[8] A A Soofi I Riaz and U Rasheed ldquoAn enhanced vigenerecipher for data securityrdquo International Journal of Scientific ampTechnology Research vol 4 no 8 pp 141ndash145 2015

[9] N Shamsuddin M Syafiqah and S Ali Pitchay ldquoImple-menting location-based cryptography on mobile applicationdesign to secure data in cloud storagerdquo Journal of PhysicsConference Series IOP Publishing vol 1551 no 1 2020

[10] L Jiao H Yonglin and D Feng ldquoStream cipher designs areviewrdquo Science China Information Sciences vol 63 no 3pp 1ndash25 2020

[11] NIST ldquoNIST cryptographic standards and guidelines devel-opment processrdquo Inside NIST vol 27 2016

[12] G Alagic J Alperin-Sheriff D Apon et al Status Report onthe First Round of the NIST Post-Quantum CryptographyStandardization Process National Institute of Standards andTechnology Gaithersburg MD USA pp 1ndash27 2019

[13] I Damaj and S Kasbah ldquoAn analysis framework for hardwareand software implementations with applications from cryp-tographyrdquo Computers amp Electrical Engineering vol 69pp 572ndash584 2018

[14] M S Croock Z A Hassan and S D Khudhur ldquoAdaptive keygeneration algorithm based on software engineering meth-odologyrdquo International Journal of Electrical and ComputerEngineering vol 11 no 1 p 589 2021

[15] K Sha T A Yang W Wei and S Davari ldquoA survey of edgecomputing-based designs for iot securityrdquo Digital Commu-nications and Networks vol 6 no 2 pp 195ndash202 2020

[16] M Yahuza M Y I B Idris A W B A Wahab et alldquoSystematic review on security and privacy requirements inedge computing state of the art and future research oppor-tunitiesrdquo IEEE Access vol 8 pp 76541ndash76567 2020

[17] A Bhowmik J Dey and S Karforma A Novel Double TierCryptographic System (nDTCS) to Reinforce Patientsrsquo Privacyin Contemporary COVID-19 Telemedicine Springer NewYork NY USA 2021

[18] N Gupta User Integrity Protection Security Model forEnhanced Telemedicine for Healthcare Networks 2020httpsosfio69jdq

[19] J Dey A Bhowmik A Sarkar S Karforma andB Chowdhury Cryptographic Engineering on COVID-19Telemedicine An Intelligent Transmission through RecurrentRelation Based Session Key Springer New York NY USA2021

[20] V Hosseinian H Ayatollahi H Haghani and E MehraeenldquoRequirements of information security in a telemedicinenetwork review of IT managersrsquo opinionrdquo Journal of Para-medical Sciences amp Rehabilitation vol 4 no 2 pp 31ndash402015

[21] D Karaoglan Altop A Levi and V Tuzcu ldquoDeriving cryp-tographic keys from physiological signalsrdquo Pervasive andMobile Computing vol 39 pp 65ndash79 2017

[22] O Uzunkol and M S Kiraz ldquoStill wrong use of pairings incryptographyrdquo Applied Mathematics and Computationvol 333 pp 467ndash479 2018

[23] S Picek D Jakobovic J F Miller L Batina and M CupicldquoCryptographic Boolean functions one output many designcriteriardquo Applied Soft Computing vol 40 pp 635ndash653 2016

[24] R Saha and G Geetha ldquoSymmetric random function gen-erator (SRFG) a novel cryptographic primitive for designingfast and robust algorithmsrdquo Chaos Solitons amp Fractalsvol 104 pp 371ndash377 2017

[25] A Achuthshankar and A Achuthshankar ldquoA novel sym-metric cryptography algorithm for fast and secure encryp-tionrdquo in Proceedings of the 2015 IEEE 9th InternationalConference on Intelligent Systems and Control (ISCO) pp 4ndash9Coimbatore India January 2015

[26] D Ott and C Peikert ldquoIdentifying research challenges in postquantum cryptography migration and cryptographic agilityrdquo2019 httparxivorgabs190907353

[27] W Liu B Ying H Yang and H Wang ldquoAccurate modelingfor predicting cryptography overheads on wireless sensornodesrdquovol 2 pp 997ndash1001 in Proceedings of the 2019 11thInternational Conference on Advanced CommunicationTechnology vol 2 pp 997ndash1001 IEEE Bangalore IndiaFebruary 2019

[28] S Feizi A Ahmadi and A Nemati ldquoA hardware imple-mentation of simon cryptography algorithmrdquo in Proceedingsof the 2014 4th International Conference on Computer andKnowledge Engineering (ICCKE) pp 245ndash250 IEEEMashhad Iran 2014

[29] E ambiraja G Ramesh and R Umarani ldquoA survey onvarious most common encryption techniquesrdquo 2012 httpwwwijarcssecom

[30] A Gupta and N K Walia ldquoCryptography Algorithms areviewrdquo International Journal of Engineering Research andDevelopment vol 2 no 2 pp 1667ndash1672 2014

[31] R Bhanot and R Hans ldquoA review and comparative analysis ofvarious encryption algorithmsrdquo International Journal of Se-curity and Its Applications vol 9 no 4 pp 289ndash306 2015

[32] S Chandra S Paira S S Alam and G Sanyal ldquoA com-parative survey of symmetric and asymmetric key

16 Computational Intelligence and Neuroscience

cryptographyrdquo in Proceedings of the 2014 InternationalConference on Electronics Communication and Computa-tional Engineering (ICECCE) pp 83ndash93 Hosur India No-vember 2014

[33] F Maqsood M Ahmed M Mumtaz and M Ali ldquoCryp-tography a comparative analysis for modern techniquesrdquoInternational Journal of Advanced Computer Science andApplications vol 8 no 6 p 6 2017

[34] S B Sadkhan and A O Salman ldquoA survey on lightweight-cryptography status and future challengesrdquo in Proceedings ofthe International Conference on Advances in SustainableEngineering and Applications pp 105ndash108 Wasit Iraq June2018

[35] N V oai ldquoCriteria and dimension reduction of linearmultiple criteria optimization problemsrdquo Journal of GlobalOptimization vol 52 no 3 pp 499ndash508 2012

[36] C E Shannon ldquoCommunication theory of secrecy systems1945rdquo MD Computing vol 15 no 1 pp 57ndash64 1998

[37] L J Fennelly M Beaudry and M A Perry Security in 2025[38] ITU Global Cybersecurity Index 2018 2019[39] ITU Global Cybersecurity Index (GCI) 2017 2017

Computational Intelligence and Neuroscience 17

σminus1 d1( 1113857 σ3 d2( 1113857⟶ 1 σ3 d2( 1113857 σ0 d3( 1113857⟶ 0 σ0 d3( 1113857 σ1 d4( 1113857⟶ minus 1

σ1 d4( 1113857 σ3 d5( 1113857⟶ 1 σ3 d5( 1113857 σ3 d6( 1113857⟶ 3 σ3 d6( 1113857 σ1 d7( 1113857⟶ 1

σ1 d7( 1113857 σ3 d8( 1113857⟶ 1 σ3 d8( 1113857 σ2 d9( 1113857⟶ 2 σ2 d9( 1113857 σ2 d10( 1113857⟶ 2

σ2 d10( 1113857 σ2 d11( 1113857⟶ 2 σ3 d11( 1113857 σ1 d12( 1113857⟶ 1 σ1 d12( 1113857 σ0 d13( 1113857⟶ 0

(4)

e maximum link dimension is 3 indicating the re-lationship between the diagnosis classes

331 Calculating the Structure Vectors As mentioned inthe definitions the vector Qq is a simplification basiscreated to eliminate the additional impacts in the equiv-alent simplex sets e maximum complex dimension withα70 7 is 3 erefore the first structure vector based onthe output is

Dimension 3 2 1 0 (5)

e second structure vector P is

dimensions 3 2 1 0

P Pdim3 Pdim2 Pdim1 Pdim0( 1113857

P 4 7 10 12( 1113857

(6)

where Pq is the number simplexes greater than or equal to qin the set K in which P is the number of simplex linkrepetitions (support indices) in the quadruple design andimplementation steps of the cryptography algorithms Basedon the values of these two vectors it is seen that the rela-tionship between the support indices and the quadrupledesign and implementation steps of the cryptography al-gorithms is high is issue indicates the role of supportcomponents in the design and implementation of the

cryptography algorithms which should be consideredseriously

332 Calculating the Obstruction or Flexibility VectorQlowastK represents the number of structural obstructions forsimplex interactions in dimension k

Qlowast

Q minus I⟶ Qlowast

1 1 1 11113858 1113859 minus 1 1 1 11113858 1113859⟶ Qlowast

0 0 0 01113858 1113859

(7)

As can be seen there is no obstruction in any of thecommunication levels indicating that there is a significantrelationship between the support components at eachequivalence class

333 Calculating Irregularity e value of (eccprime(σ)) iscalculated using the Chinese method e results for α70

7 are shown in Table 7 e calculated irregularity valueshows that the indices d2 (structureorganization) d5 (hu-man resourceseducation) d6 (research and development)and d8 (management) affect other indices

334 Calculating Complexity Also the results ofQ-analysiscan be used to describe structure complexity According toequations (4)ndash(9) and for α = 7 the complexity measure is

-1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1

3 0 1 3 3 1 3 2 2 2 1

0 -1 0 0 0 0 0 0 0 0

1 1 1 0 1 1 1 1 0

3 3 1 3 2 2 2 1

3 1 3 2 2 2 1

3 2 2 2 1

2 2 2 1

2 2 1

2 1

1

1 1 1 1 1 1

-1

-1

0

0

0

0

0

0

0

0

0

0

0q Q3210

11

11

SETSX2 X5 X6 X8X2 X5 X6 X8 X9 X10 X11X2 X4 X5 X6 X7 X8 X9 X10 X11 X12X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13

Figure 4 Implementation results of the model for support components of the cryptography algorithmsrsquo design with α 70

8 Computational Intelligence and Neuroscience

Q 1 1 1 1( 1113857

ψ(K) 2(1 + 2 + 3 + 4)

(4lowast 5)1113890 1113891 1

(8)

Since in the above model there is no obstruction amongthe component of the equivalent class it was expected thatthe complexity of the support components is not high andthe obtained complexity index of 1 verifies this expectatione system complexity for different alpha-cuts is shown inFigure 5

335 Ranking the Support Components of the Design andImplementation of the Cryptography Algorithms e resultsof using A-analysis are shown in Table 7 e connectionstrength of the factors in one group is specified with alpha-cut erefore the support components are grouped in 5levels Each level describes the priority and importance of thegroup in developing the cryptography algorithms In Fig-ure 6 the ranking pyramid of the support components usingQ-analysis is shown To allocate proper resources thecomponents existing in higher levels of the pyramid (Fig-ure 6) are of higher priority

336 Validation of the Results In this section the resultsof the cybernetic model and Q-analysis for supportcomponentsrsquo ranking are compared with the results re-ported in the global cybersecurity index (GCI) in 20152017 and 2018 presented by ITU [36ndash39] e GCI reportsare focused on five indices including ldquolegal cases orga-nization necessities technical issues capacity buildingand cooperationrdquo and the subindices include legaltechnical organization capacity building and coopera-tion According to the presented indices and subindices itis clear that the ldquorulesrdquo ldquostandardrdquo ldquoresearch and de-velopmentrdquo ldquoeducationrdquo and ldquomanagementrdquo are ofhigher priority in security establishment Although ourresearch is more skilled and detailed compared to the GCIreports but the results verify our findings Also Table 8shows the ranking of the support components obtainedusing Q-analysis

337 Executing the Model and Analyzing the Results of theObjectivesrsquo Impact on Design Steps of the Cryptography Al-gorithmsrsquo Matrix In this section the role of the sevencomponents on the quadruple design and implementationsteps of the cryptography algorithm is analyzed with Q-analysis Using the Q-analysis method and the 7lowast4 matrixobtained from the relationship of the cryptography algo-rithmsrsquo design objectives on their quadruple steps theirindices are ranked

338 Calculating the Incidence Matrix and Executing theModel First the incidence matrix is obtained for the datamatrix shown in Figure 2 It can be seen in Table 9 that eachelement of the two sets represents which indice e inci-dence matrix calculated from the data matrix for α5 40 is

given in Table 10 e results of implementing the model forα5 40 are given in Figure 7

4 Geometric Representation

In the sample with α5 40 dis are

d1 c1 c2 c3 c41113864 1113865

d2

d3 c1 c3 c41113864 1113865

d4 c3 c41113864 1113865

d5

d6 c1 c2 c3 c41113864 1113865

d7

(9)

e simplexes of σp(di) are also

σ3 d1( 1113857σminus1 d2( 1113857σ2 d3( 1113857σ1 d4( 1113857σminus1 d5( 1113857σ3 d6( 1113857σminus1 d7( 1113857

(10)

erefore the complex dimension is 3 In other wordsthe discriminant classes d1 (security) and d6 (speed) have thelargest dimension

41 Calculating the Structure Vectors As mentioned thevector Qq is a simplification basis to eliminate the additionaleffects in the set of equivalent simplexes e maximumcomplex dimension for α5 40 is 3 erefore the firststructure vector based on the output is

Dimension 3 2 1 1

Q 1 1 1 1( 1113857

Dimension 3 2 1 0

P 2 3 4 7( 1113857

(11)

e second structure vector P is

P 2 3 4 7( 1113857 (12)

42 Calculating the Obstruction Vector or InflexibilityQlowastK represents the number of structural obstructions forsimplex interactions in dimension k which is as follows forα5 40

1 1 1 1 1332

005

115

225

com

plex

ity

alpha cutα=5 α=6 α=7 α=8 α=9 α=10

Figure 5 System complexity of the support indices of design andimplementation of the cryptography algorithms for different alpha-cuts

Computational Intelligence and Neuroscience 9

Qlowast

Q minus I⟶ Qlowast

2 1 1 31113858 1113859 minus 1 1 1 11113858 1113859⟶ Qlowast

0 0 0 01113858 1113859

(13)

According to the obtained values it can be concludedthat there are no structural obstructions among the main

design indices of the cryptography algorithms at ismultiple objectives are considered simultaneously by thecryptography algorithm designers

Table 7 e irregularity of the data matrix parameters for α= 7

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d2 d5 d6 d8 qi 3 2 1 364 3 061d9 d10d11 qi 2 1 064 3 011d4 d7 d12 qi 1 014 3 002

human resourceRampD

rules FAVA standard

management infrastructure

organizationstructure equipmentinternational and public relations finanacial

resources and material

Figure 6 e ranking pyramid of the support components using Q-analysis

Table 8 Ranking of the support components using Q-analysis

Relationship of the support components with the design and implementation of the cryptography algorithms (q 0)No relationship α 0 complete relationship α 100

Human resource RampD rules FAVA and standard α100 10Management and infrastructure α90 9Organizationstructure and equipment α80 8International and public relations financial resources and material α70 7Culture α60 6All components α60 5

Table 9 Set of dis and cis the objectivesrsquo indices

eoretical basis C2 Reliability d7 Scalability d5 Resources d3 Security d1Implementation C3 Application C1 Speed d6 Flexibility d4 Simplicity d2Evaluation C4

Table 10 e incidence matrix of the design objectives of thecryptography algorithms with α= 40

c1 c2 c3 c4

d1 1 1 1 1d2 0 0 0 0d3 1 0 1 1d4 0 0 1 1d5 0 0 0 0d6 1 1 1 1d7 0 0 0 0

-1 -1 -1 -1 -1

2 1 -1 2

1 -1 1

-1 -1

3

3 -1 2 1 -1 3 -1

-1

-1

-1

-1

-1

-1q Q3210

11

11

SETSX1X6X1X3X6X1X3X4X6X1X3X4X6

Figure 7 e results of executing the model for design objectivesrsquoindices with α 40

10 Computational Intelligence and Neuroscience

43 Calculating Irregularity e results of applying theChinese method (ecclsquo(σ)) for calculating irregularity forα5 40 are shown in Table 11 e calculated irregu-larity value shows that indices d1 (security) and d6(speed) have received more attention compared to otherindices

44 Calculating Complexity Structural complexity for asample with α5 40 is

dimensions 3 2 1 0

Q Qdim3 Qdim2 Qdim1 Qdim0( 1113857

Q 1 1 1 1( 1113857

ψ(K) 2(1 + 2 + 3 + 4)

(4lowast 5)1113890 1113891 1

(14)

Since in the above model there was no obstructionamong the components of the equivalent classes in any of thecommunication levels it was expected that there is not ahigh complexity among objectives of the cryptography al-gorithm design and the complexity index of 1 verified thisexpectation e system complexity for different alpha-cutsis shown in Figure 8 As can be seen the system complexity is1 for all significant values of alpha

45 Prioritizing the Parameters of the Main Objectives of theCryptography Algorithm Design Using Q-AnalysisAccording to the analysis results the parameters can beclassified into 5 levelsese levels are shown in Table 12 andFigure 9 Each level indicates a priority and importance ofthe group in the development of cryptography algorithmsTo allocate proper resources the components at the higherlevels of the pyramid are of higher priority As can be seenthree objectives of security speed and optimal usage ofresources have the highest priority for the design of cryp-tography algorithms

46 Executing the Model and Analyzing the Results for theCryptography Algorithm Implementation Techniques einteraction matrix between the 29 extracted techniques andthe sevenmain objectives is shown in Figure 10e purposeof this section is to rank the cryptography algorithmimplementation techniques to achieve the goals of interest

47Calculating the IncidenceMatrixandExecuting theModele data matrix A is comprised of two sets e set Drepresents the employed techniques and the set C representsthe seven main design objectives (Table 13)

D d1 d2 d291113864 1113865

C c1 c2 c71113864 1113865(15)

Some parts of the incidence matrix calculated from thedata matrix A for α3 are shown in Table 14 and the resultsof theQ-analysis model for α3 24 are shown in Figure 11

48 Geometric Representation In the sample with α3 24the dis are

d1 c11113864 1113865

d2

d3

d4

d5

d6

d7

d8

d9

d10

d11 c11113864 1113865

d12

d13 c51113864 1113865

d14

d15 c1 c4 c51113864 1113865

d16

d17

d18

d19 c11113864 1113865

d20

d21

d22 c31113864 1113865

d23

d24

d25

d26

d27 c11113864 1113865

d28

d29

(16)

e simplexes of σp(di) are

Computational Intelligence and Neuroscience 11

σ0 d1( 1113857

σminus1 d2( 1113857

σminus1 d3( 1113857

σminus1 d4( 1113857

σminus1 d5( 1113857

σminus1 d6( 1113857

σminus1 d7( 1113857

σminus1 d8( 1113857

σminus1 d9( 1113857

σminus1 d10( 1113857

σ0 d11( 1113857

σminus1 d12( 1113857

σ0 d13( 1113857

σminus1 d14( 1113857

σ2 d15( 1113857

σminus1 d16( 1113857

σminus1 d17( 1113857

σminus1 d18( 1113857

σ0 d19( 1113857

σminus1 d20( 1113857

σminus1 d21( 1113857

σ0 d22( 1113857

σminus1 d23( 1113857

σminus1 d24( 1113857

σminus1 d25( 1113857

σminus1 d26( 1113857

σ0 d27( 1113857

σminus1 d28( 1113857

σminus1 d29( 1113857

(17)

erefore the complex dimension is 2 In other wordsthe discriminant class d15 (basic sciences) has the largestdimension

49 Calculating the Dimensions and the Q-Link In the alphadefined for cryptography algorithm implementation tech-niques to achieve the defined goals the obtained q-linkshows that these techniques are relatively independent al-though there is a weak relationship between some tech-niques Q-link in the samples with α3 24 between eachtwo subsequent di is as follows

σ0 d1( 1113857 σminus1 d2( 1113857⟶ minus 1

σminus1 d2( 1113857 σminus1 d3( 1113857⟶ minus 1

σ0 d3( 1113857 σminus1 d4( 1113857⟶ minus 1

σminus1 d4( 1113857 σminus1 d5( 1113857⟶ minus 1

σminus1 d5( 1113857 σminus1 d6( 1113857⟶ minus 1

σminus1 d6( 1113857 σminus1 d7( 1113857⟶ minus 1

σminus1 d7( 1113857 σminus1 d8( 1113857⟶ minus 1

σminus1 d8( 1113857 σminus1 d9( 1113857⟶ minus 1

σminus1 d9( 1113857 σminus1 d10( 1113857⟶ minus 1

σminus1 d10( 1113857 σ0 d11( 1113857⟶ minus 1

σ0 d11( 1113857 σminus1 d12( 1113857⟶ minus 1

σminus1 d12( 1113857 σ0 d13( 1113857⟶ minus 1

σ0 d13( 1113857 σminus1 d14( 1113857⟶ minus 1

σminus1 d14( 1113857 σ2 d15( 1113857⟶ minus 1

σ2 d15( 1113857 σminus1 d16( 1113857⟶ minus 1

σminus1 d16( 1113857 σminus1 d17( 1113857⟶ minus 1

σminus1 d17( 1113857 σminus1 d18( 1113857⟶ minus 1

σminus1 d18( 1113857 σ0 d19( 1113857⟶ minus 1

σ0 d19( 1113857 σminus1 d20( 1113857⟶ minus 1

σminus1 d20( 1113857 σminus1 d21( 1113857⟶ minus 1

σminus1 d21( 1113857 σ0 d22( 1113857⟶ minus 1

σ0 d22( 1113857 σminus1 d23( 1113857⟶ minus 1

σminus1 d23( 1113857 σminus1 d24( 1113857⟶ minus 1

σminus1 d24( 1113857 σminus1 d25( 1113857⟶ minus 1

σminus1 d25( 1113857 σminus1 d26( 1113857⟶ minus 1

σminus1 d26( 1113857 σ0 d27( 1113857⟶ minus 1

σ0 d27( 1113857 σminus1 d28( 1113857⟶ minus 1

σminus1 d28( 1113857 σminus1 d29( 1113857⟶ minus 1

(18)

Table 11 Irregularity of the cryptography objectivesrsquo parameters in the data matrix A for α5

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d1 and d6 qi 3 2 and 1 433 3 072d3 qi 2 and 1 133 3 026d4 qi 1 033 3 006

11117

1 1 1

09095

1105

11115

12

com

plex

ity

alpha cutα=2 α=3 α=5 α=15 α=25

Figure 8 System complexity of the cryptography algorithm designobjectives for different alpha-cuts

12 Computational Intelligence and Neuroscience

410 Calculating the Structure Vectors erefore the firststructure vector based on the software output is

Q 1 1 2( 1113857 (19)

e second structure vector P is

P 1 1 29( 1113857 (20)

Since the large values of P of higher dimensions dem-onstrate more links the second structure vector calculatedfor the alpha of interest shows that the relationship betweenthe cryptography algorithm implementation techniques isminimum

411 Calculating the Obstruction or Inflexibility Vectore obstruction vector (Qlowast) for a sample with α3 24 is

Table 12 Ranking the objectivesrsquo parameters considering different cuts in Q-analysis

e equivalence class for the parameters with minimum relation level (q 0) αSecurity α50 244Speed α30 244Resources α15 122Flexibility α5 40Simplicity-scalability α2 16(All parameters)

securityspeed

optimal usage of resources

flexibility

simplicity-scalability

Figure 9 Prioritizing the design and implementation objectives of the cryptography algorithms using Q-analysis

securityspeed

optimal usage of resources

flexibility

simplicity-scalability-reliability

Figure 10 Prioritizing the design and implementation objectives of the cryptography algorithms using Q-analysis

Table 13 Set of dis and cis the techniques employed for implementing cryptography algorithms

Flexibility C4 Fuzzy logic d25 Storage space d17 Cycle d9 Avalanche effect d1Scalability C5 Software d26 Clustering d18 Multiple step d10 Digital signature d2Speed C6 Steganography d27 Key d19 Hybrid method d11 Block size d3Reliability C7 Music harmony d28 Graph d20 Hardware d12 Image sharing d4

Artificial intelligence d29 Characteristic oriented d21 Hardware d13 Parallel processing d5Security C1 Energy consumption d22 Occupation area d14 reshold technique d6Simplicity C2 Bandwidth consumption d23 Basic science d15 Data mining d7Resources C3 Memory consumption d24 Compression d16 Binary tree d8

Table 14 e incidence matrix of the cryptography algorithmimplementation techniques with α 24

C1 C2 C3 C4 C5 C6 C7

d1 1 0 0 0 0 0 0d2 0 0 0 0 0 0 0d3 0 0 0 0 0 0 0d4 0 0 0 0 0 0 0d5 0 0 0 0 0 0 0d6 0 0 0 0 0 0 0d7 0 0 0 0 0 0 0d8 0 0 0 0 0 0 0d9 0 0 0 0 0 0 0d10 0 0 0 0 0 0 0d11 1 0 0 0 0 0 0d12 0 0 0 0 0 0 0

Computational Intelligence and Neuroscience 13

Qlowast

Q minus I⟶ Qlowast

1 1 21113858 1113859 minus 1 1 11113858 1113859⟶ Qlowast

0 0 11113858 1113859

(21)

e value of QlowastK represents the number of structurallimitations or obstructions for the cryptography techniquesrsquointeraction at dimension k As can be seen from the cal-culated obstruction vector there is a significant relationshipat some levels

412 Calculating Irregularity Irregularity is the integrationdegree of a cryptography method in the total complex Mea-suring irregularity (eccrsquo) for α3 24 is shown in Table 15 Ascan be seen irregularity for the discriminant class d15 indicatesthat this technique is isolated from other techniques

413 Calculating Complexity Structure complexity for asample with α3 24 is

Q 1 1 2( 1113857

ψ(K) 2(2 + 2 + 3)

(3lowast 4)1113890 1113891 117

(22)

e system complexity for different values of alpha isshown in Figure 12 As can be seen the total complexity ofthe system for large values alpha tends to stability

414 Ranking the Cryptography Algorithm ImplementationTechniques ese techniques can be classified into 5 levelsusing the analysis of the obtained results ese levels can beseen in Table 16 and Figure 13 e link power of the factorsin one group is specified with alpha Each level describespriority and importance of the group in the development ofcryptography algorithms Accordingly six methods of basicscience key management using hardware methods usingsteganography Avalanche effect and using hybrid method

q Q21o

112

SETSX15X15X1X11X15X19X27X13 amp X22

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-10-1-1-10-1-1-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-10-1-1-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

2-10-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1

-1-1-1-1-1-1

-1-1-1-1-1

-1-1-1-1

-1-1-1

-1-10

Figure 11 Results of executing the model for cryptography algorithm implementation techniques with α3 24

14 Computational Intelligence and Neuroscience

are the most important cryptography algorithm imple-mentation techniques to achieve the main objectives

5 Conclusion

In this study the design and implementation model of thecryptography algorithms is designed in three levels and itseffective components are extracted To organize the com-ponents in addition to elimination of the additional com-ponents through the measure reduction algorithm a properclassification is applied to examine the mutual effects Afterclassification three 13lowast 4 7lowast 4 and 29lowast 7 matrices areconstituted and the model is implemented on these threematrices To implement the designed model Q-analysis isused Accordingly for support indices five indices of highpriority include human resources research and develop-ment management organizationstructure and equipmentFor the algorithm design objectives index five high priorityindices include security speed optimal usage of resources

and simplicity For the indices related to implementationtechniques of the cryptography algorithms the most appliedtechniques for achieving the determined objectives includeusing basic science hardware and software methods usingkeymanagement hybrid method steganography Avalancheeffect and using artificial intelligence In the future work wewill consider the following cases

A plan should be formulated according to the prioritiesAccording to the outputs of this study and the presented pri-orities the following topics can be investigated in future studies

(1) Presenting a comprehensive model for generatingvarious cryptography algorithms based on the pri-orities of interest

(2) Examining and formulating a model about the roleof basic science for design and implementation ofcryptography algorithms considering the significantrole of ldquobasic sciencerdquo in implementation of thecryptography algorithms and about the role ofschools and universities

(3) Presenting a model for design and implementationof cryptography algorithms in the IoT infrastructurewith optimal resource usage

Data Availability

e data used to support the findings of the study areavailable from the corresponding author upon request

Conflicts of Interest

e authors declare that there are no conflicts of interestregarding the publication of this article

Table 15 Irregularity of the cryptography algorithm implementation techniques for α3 24

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d15 qi 2 1 3 2 1- qi 1 1 2 033

11117

1 1 1

091

1112

com

plex

ity

alpha-cutα=2 α=3 α=5 α=15 α=25

Figure 12 System complexity of cryptography algorithm implementation techniques for different alpha-cuts

Table 16 Ranking the implementation techniques of the cryptography algorithms considering various alpha-cuts using Q-analysis

Equivalence classes for the parameters with minimum relationship level (q 0) αBasic sciences α20 163Key α10 81Hardware and steganography α5 40Avalanche effect and hybrid method α3 24Artificial intelligence software occupation area energy consumption memory consumption α2 16

basic sciences

key

Avalanche effect hybrid method

hardware steganography

artificial intelligence soware occupationarea energy consumption memory consumption

Figure 13 Prioritizing the implementation techniques of thecryptography algorithms using Q-analysis

Computational Intelligence and Neuroscience 15

Acknowledgments

is work was partially supported by Iran National ScienceFoundation (INSF) under contract No 9653979

References

[1] D Nilesh and M Nagle ldquoe new cryptography algorithmwith high throughputrdquo in Proceedings of the 2014 Interna-tional Conference on Computer Communication and Infor-matics pp 3ndash7 Coimbatore India January 2014

[2] S Ravi P Kocher R Lee G McGraw and A RaghunathanldquoSecurity as a new dimension in embedded system designrdquo inProceedings of the 41st annual conference on Design auto-mation-DAC lsquo04 pp 753ndash760 San Diego CA USA March2004

[3] P Kuppuswamy and S Q Y A Khalidi ldquoHybrid encryptiondecryption technique using new public key and symmetric keyalgorithmrdquo International Journal of Information and Com-puter Security vol 6 no 4 pp 372ndash382 2014

[4] S Bhat and V Kapoor ldquoSecure and efficient data privacyauthentication and integrity schemes using hybrid cryptog-raphyrdquo International Conference on Advanced ComputingNetworking and Informatics vol 870 pp 279ndash285 2019

[5] D Ott and C Peikert ldquoAnd other workshop participantsldquoidentifying research challenges in post quantum cryptog-raphy migration and cryptographic agilityrdquo 2019 httparxivorgabs190907353

[6] P Manickam K Shankar E Perumal M Ilayaraja andK Sathesh Kumar ldquoSecure data transmission through reliablevehicles in VANET using optimal lightweight cryptographyrdquoin Advanced Sciences and Technologies for Security Applica-tions pp 193ndash204 Springer New York NY USA 2019

[7] S Prakash and A Rajput ldquoHybrid cryptography for securedata communication in wireless sensor networksrdquo Advancesin Intelligent Systems and Computing vol 696 pp 589ndash5992018

[8] A A Soofi I Riaz and U Rasheed ldquoAn enhanced vigenerecipher for data securityrdquo International Journal of Scientific ampTechnology Research vol 4 no 8 pp 141ndash145 2015

[9] N Shamsuddin M Syafiqah and S Ali Pitchay ldquoImple-menting location-based cryptography on mobile applicationdesign to secure data in cloud storagerdquo Journal of PhysicsConference Series IOP Publishing vol 1551 no 1 2020

[10] L Jiao H Yonglin and D Feng ldquoStream cipher designs areviewrdquo Science China Information Sciences vol 63 no 3pp 1ndash25 2020

[11] NIST ldquoNIST cryptographic standards and guidelines devel-opment processrdquo Inside NIST vol 27 2016

[12] G Alagic J Alperin-Sheriff D Apon et al Status Report onthe First Round of the NIST Post-Quantum CryptographyStandardization Process National Institute of Standards andTechnology Gaithersburg MD USA pp 1ndash27 2019

[13] I Damaj and S Kasbah ldquoAn analysis framework for hardwareand software implementations with applications from cryp-tographyrdquo Computers amp Electrical Engineering vol 69pp 572ndash584 2018

[14] M S Croock Z A Hassan and S D Khudhur ldquoAdaptive keygeneration algorithm based on software engineering meth-odologyrdquo International Journal of Electrical and ComputerEngineering vol 11 no 1 p 589 2021

[15] K Sha T A Yang W Wei and S Davari ldquoA survey of edgecomputing-based designs for iot securityrdquo Digital Commu-nications and Networks vol 6 no 2 pp 195ndash202 2020

[16] M Yahuza M Y I B Idris A W B A Wahab et alldquoSystematic review on security and privacy requirements inedge computing state of the art and future research oppor-tunitiesrdquo IEEE Access vol 8 pp 76541ndash76567 2020

[17] A Bhowmik J Dey and S Karforma A Novel Double TierCryptographic System (nDTCS) to Reinforce Patientsrsquo Privacyin Contemporary COVID-19 Telemedicine Springer NewYork NY USA 2021

[18] N Gupta User Integrity Protection Security Model forEnhanced Telemedicine for Healthcare Networks 2020httpsosfio69jdq

[19] J Dey A Bhowmik A Sarkar S Karforma andB Chowdhury Cryptographic Engineering on COVID-19Telemedicine An Intelligent Transmission through RecurrentRelation Based Session Key Springer New York NY USA2021

[20] V Hosseinian H Ayatollahi H Haghani and E MehraeenldquoRequirements of information security in a telemedicinenetwork review of IT managersrsquo opinionrdquo Journal of Para-medical Sciences amp Rehabilitation vol 4 no 2 pp 31ndash402015

[21] D Karaoglan Altop A Levi and V Tuzcu ldquoDeriving cryp-tographic keys from physiological signalsrdquo Pervasive andMobile Computing vol 39 pp 65ndash79 2017

[22] O Uzunkol and M S Kiraz ldquoStill wrong use of pairings incryptographyrdquo Applied Mathematics and Computationvol 333 pp 467ndash479 2018

[23] S Picek D Jakobovic J F Miller L Batina and M CupicldquoCryptographic Boolean functions one output many designcriteriardquo Applied Soft Computing vol 40 pp 635ndash653 2016

[24] R Saha and G Geetha ldquoSymmetric random function gen-erator (SRFG) a novel cryptographic primitive for designingfast and robust algorithmsrdquo Chaos Solitons amp Fractalsvol 104 pp 371ndash377 2017

[25] A Achuthshankar and A Achuthshankar ldquoA novel sym-metric cryptography algorithm for fast and secure encryp-tionrdquo in Proceedings of the 2015 IEEE 9th InternationalConference on Intelligent Systems and Control (ISCO) pp 4ndash9Coimbatore India January 2015

[26] D Ott and C Peikert ldquoIdentifying research challenges in postquantum cryptography migration and cryptographic agilityrdquo2019 httparxivorgabs190907353

[27] W Liu B Ying H Yang and H Wang ldquoAccurate modelingfor predicting cryptography overheads on wireless sensornodesrdquovol 2 pp 997ndash1001 in Proceedings of the 2019 11thInternational Conference on Advanced CommunicationTechnology vol 2 pp 997ndash1001 IEEE Bangalore IndiaFebruary 2019

[28] S Feizi A Ahmadi and A Nemati ldquoA hardware imple-mentation of simon cryptography algorithmrdquo in Proceedingsof the 2014 4th International Conference on Computer andKnowledge Engineering (ICCKE) pp 245ndash250 IEEEMashhad Iran 2014

[29] E ambiraja G Ramesh and R Umarani ldquoA survey onvarious most common encryption techniquesrdquo 2012 httpwwwijarcssecom

[30] A Gupta and N K Walia ldquoCryptography Algorithms areviewrdquo International Journal of Engineering Research andDevelopment vol 2 no 2 pp 1667ndash1672 2014

[31] R Bhanot and R Hans ldquoA review and comparative analysis ofvarious encryption algorithmsrdquo International Journal of Se-curity and Its Applications vol 9 no 4 pp 289ndash306 2015

[32] S Chandra S Paira S S Alam and G Sanyal ldquoA com-parative survey of symmetric and asymmetric key

16 Computational Intelligence and Neuroscience

cryptographyrdquo in Proceedings of the 2014 InternationalConference on Electronics Communication and Computa-tional Engineering (ICECCE) pp 83ndash93 Hosur India No-vember 2014

[33] F Maqsood M Ahmed M Mumtaz and M Ali ldquoCryp-tography a comparative analysis for modern techniquesrdquoInternational Journal of Advanced Computer Science andApplications vol 8 no 6 p 6 2017

[34] S B Sadkhan and A O Salman ldquoA survey on lightweight-cryptography status and future challengesrdquo in Proceedings ofthe International Conference on Advances in SustainableEngineering and Applications pp 105ndash108 Wasit Iraq June2018

[35] N V oai ldquoCriteria and dimension reduction of linearmultiple criteria optimization problemsrdquo Journal of GlobalOptimization vol 52 no 3 pp 499ndash508 2012

[36] C E Shannon ldquoCommunication theory of secrecy systems1945rdquo MD Computing vol 15 no 1 pp 57ndash64 1998

[37] L J Fennelly M Beaudry and M A Perry Security in 2025[38] ITU Global Cybersecurity Index 2018 2019[39] ITU Global Cybersecurity Index (GCI) 2017 2017

Computational Intelligence and Neuroscience 17

Q 1 1 1 1( 1113857

ψ(K) 2(1 + 2 + 3 + 4)

(4lowast 5)1113890 1113891 1

(8)

Since in the above model there is no obstruction amongthe component of the equivalent class it was expected thatthe complexity of the support components is not high andthe obtained complexity index of 1 verifies this expectatione system complexity for different alpha-cuts is shown inFigure 5

335 Ranking the Support Components of the Design andImplementation of the Cryptography Algorithms e resultsof using A-analysis are shown in Table 7 e connectionstrength of the factors in one group is specified with alpha-cut erefore the support components are grouped in 5levels Each level describes the priority and importance of thegroup in developing the cryptography algorithms In Fig-ure 6 the ranking pyramid of the support components usingQ-analysis is shown To allocate proper resources thecomponents existing in higher levels of the pyramid (Fig-ure 6) are of higher priority

336 Validation of the Results In this section the resultsof the cybernetic model and Q-analysis for supportcomponentsrsquo ranking are compared with the results re-ported in the global cybersecurity index (GCI) in 20152017 and 2018 presented by ITU [36ndash39] e GCI reportsare focused on five indices including ldquolegal cases orga-nization necessities technical issues capacity buildingand cooperationrdquo and the subindices include legaltechnical organization capacity building and coopera-tion According to the presented indices and subindices itis clear that the ldquorulesrdquo ldquostandardrdquo ldquoresearch and de-velopmentrdquo ldquoeducationrdquo and ldquomanagementrdquo are ofhigher priority in security establishment Although ourresearch is more skilled and detailed compared to the GCIreports but the results verify our findings Also Table 8shows the ranking of the support components obtainedusing Q-analysis

337 Executing the Model and Analyzing the Results of theObjectivesrsquo Impact on Design Steps of the Cryptography Al-gorithmsrsquo Matrix In this section the role of the sevencomponents on the quadruple design and implementationsteps of the cryptography algorithm is analyzed with Q-analysis Using the Q-analysis method and the 7lowast4 matrixobtained from the relationship of the cryptography algo-rithmsrsquo design objectives on their quadruple steps theirindices are ranked

338 Calculating the Incidence Matrix and Executing theModel First the incidence matrix is obtained for the datamatrix shown in Figure 2 It can be seen in Table 9 that eachelement of the two sets represents which indice e inci-dence matrix calculated from the data matrix for α5 40 is

given in Table 10 e results of implementing the model forα5 40 are given in Figure 7

4 Geometric Representation

In the sample with α5 40 dis are

d1 c1 c2 c3 c41113864 1113865

d2

d3 c1 c3 c41113864 1113865

d4 c3 c41113864 1113865

d5

d6 c1 c2 c3 c41113864 1113865

d7

(9)

e simplexes of σp(di) are also

σ3 d1( 1113857σminus1 d2( 1113857σ2 d3( 1113857σ1 d4( 1113857σminus1 d5( 1113857σ3 d6( 1113857σminus1 d7( 1113857

(10)

erefore the complex dimension is 3 In other wordsthe discriminant classes d1 (security) and d6 (speed) have thelargest dimension

41 Calculating the Structure Vectors As mentioned thevector Qq is a simplification basis to eliminate the additionaleffects in the set of equivalent simplexes e maximumcomplex dimension for α5 40 is 3 erefore the firststructure vector based on the output is

Dimension 3 2 1 1

Q 1 1 1 1( 1113857

Dimension 3 2 1 0

P 2 3 4 7( 1113857

(11)

e second structure vector P is

P 2 3 4 7( 1113857 (12)

42 Calculating the Obstruction Vector or InflexibilityQlowastK represents the number of structural obstructions forsimplex interactions in dimension k which is as follows forα5 40

1 1 1 1 1332

005

115

225

com

plex

ity

alpha cutα=5 α=6 α=7 α=8 α=9 α=10

Figure 5 System complexity of the support indices of design andimplementation of the cryptography algorithms for different alpha-cuts

Computational Intelligence and Neuroscience 9

Qlowast

Q minus I⟶ Qlowast

2 1 1 31113858 1113859 minus 1 1 1 11113858 1113859⟶ Qlowast

0 0 0 01113858 1113859

(13)

According to the obtained values it can be concludedthat there are no structural obstructions among the main

design indices of the cryptography algorithms at ismultiple objectives are considered simultaneously by thecryptography algorithm designers

Table 7 e irregularity of the data matrix parameters for α= 7

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d2 d5 d6 d8 qi 3 2 1 364 3 061d9 d10d11 qi 2 1 064 3 011d4 d7 d12 qi 1 014 3 002

human resourceRampD

rules FAVA standard

management infrastructure

organizationstructure equipmentinternational and public relations finanacial

resources and material

Figure 6 e ranking pyramid of the support components using Q-analysis

Table 8 Ranking of the support components using Q-analysis

Relationship of the support components with the design and implementation of the cryptography algorithms (q 0)No relationship α 0 complete relationship α 100

Human resource RampD rules FAVA and standard α100 10Management and infrastructure α90 9Organizationstructure and equipment α80 8International and public relations financial resources and material α70 7Culture α60 6All components α60 5

Table 9 Set of dis and cis the objectivesrsquo indices

eoretical basis C2 Reliability d7 Scalability d5 Resources d3 Security d1Implementation C3 Application C1 Speed d6 Flexibility d4 Simplicity d2Evaluation C4

Table 10 e incidence matrix of the design objectives of thecryptography algorithms with α= 40

c1 c2 c3 c4

d1 1 1 1 1d2 0 0 0 0d3 1 0 1 1d4 0 0 1 1d5 0 0 0 0d6 1 1 1 1d7 0 0 0 0

-1 -1 -1 -1 -1

2 1 -1 2

1 -1 1

-1 -1

3

3 -1 2 1 -1 3 -1

-1

-1

-1

-1

-1

-1q Q3210

11

11

SETSX1X6X1X3X6X1X3X4X6X1X3X4X6

Figure 7 e results of executing the model for design objectivesrsquoindices with α 40

10 Computational Intelligence and Neuroscience

43 Calculating Irregularity e results of applying theChinese method (ecclsquo(σ)) for calculating irregularity forα5 40 are shown in Table 11 e calculated irregu-larity value shows that indices d1 (security) and d6(speed) have received more attention compared to otherindices

44 Calculating Complexity Structural complexity for asample with α5 40 is

dimensions 3 2 1 0

Q Qdim3 Qdim2 Qdim1 Qdim0( 1113857

Q 1 1 1 1( 1113857

ψ(K) 2(1 + 2 + 3 + 4)

(4lowast 5)1113890 1113891 1

(14)

Since in the above model there was no obstructionamong the components of the equivalent classes in any of thecommunication levels it was expected that there is not ahigh complexity among objectives of the cryptography al-gorithm design and the complexity index of 1 verified thisexpectation e system complexity for different alpha-cutsis shown in Figure 8 As can be seen the system complexity is1 for all significant values of alpha

45 Prioritizing the Parameters of the Main Objectives of theCryptography Algorithm Design Using Q-AnalysisAccording to the analysis results the parameters can beclassified into 5 levelsese levels are shown in Table 12 andFigure 9 Each level indicates a priority and importance ofthe group in the development of cryptography algorithmsTo allocate proper resources the components at the higherlevels of the pyramid are of higher priority As can be seenthree objectives of security speed and optimal usage ofresources have the highest priority for the design of cryp-tography algorithms

46 Executing the Model and Analyzing the Results for theCryptography Algorithm Implementation Techniques einteraction matrix between the 29 extracted techniques andthe sevenmain objectives is shown in Figure 10e purposeof this section is to rank the cryptography algorithmimplementation techniques to achieve the goals of interest

47Calculating the IncidenceMatrixandExecuting theModele data matrix A is comprised of two sets e set Drepresents the employed techniques and the set C representsthe seven main design objectives (Table 13)

D d1 d2 d291113864 1113865

C c1 c2 c71113864 1113865(15)

Some parts of the incidence matrix calculated from thedata matrix A for α3 are shown in Table 14 and the resultsof theQ-analysis model for α3 24 are shown in Figure 11

48 Geometric Representation In the sample with α3 24the dis are

d1 c11113864 1113865

d2

d3

d4

d5

d6

d7

d8

d9

d10

d11 c11113864 1113865

d12

d13 c51113864 1113865

d14

d15 c1 c4 c51113864 1113865

d16

d17

d18

d19 c11113864 1113865

d20

d21

d22 c31113864 1113865

d23

d24

d25

d26

d27 c11113864 1113865

d28

d29

(16)

e simplexes of σp(di) are

Computational Intelligence and Neuroscience 11

σ0 d1( 1113857

σminus1 d2( 1113857

σminus1 d3( 1113857

σminus1 d4( 1113857

σminus1 d5( 1113857

σminus1 d6( 1113857

σminus1 d7( 1113857

σminus1 d8( 1113857

σminus1 d9( 1113857

σminus1 d10( 1113857

σ0 d11( 1113857

σminus1 d12( 1113857

σ0 d13( 1113857

σminus1 d14( 1113857

σ2 d15( 1113857

σminus1 d16( 1113857

σminus1 d17( 1113857

σminus1 d18( 1113857

σ0 d19( 1113857

σminus1 d20( 1113857

σminus1 d21( 1113857

σ0 d22( 1113857

σminus1 d23( 1113857

σminus1 d24( 1113857

σminus1 d25( 1113857

σminus1 d26( 1113857

σ0 d27( 1113857

σminus1 d28( 1113857

σminus1 d29( 1113857

(17)

erefore the complex dimension is 2 In other wordsthe discriminant class d15 (basic sciences) has the largestdimension

49 Calculating the Dimensions and the Q-Link In the alphadefined for cryptography algorithm implementation tech-niques to achieve the defined goals the obtained q-linkshows that these techniques are relatively independent al-though there is a weak relationship between some tech-niques Q-link in the samples with α3 24 between eachtwo subsequent di is as follows

σ0 d1( 1113857 σminus1 d2( 1113857⟶ minus 1

σminus1 d2( 1113857 σminus1 d3( 1113857⟶ minus 1

σ0 d3( 1113857 σminus1 d4( 1113857⟶ minus 1

σminus1 d4( 1113857 σminus1 d5( 1113857⟶ minus 1

σminus1 d5( 1113857 σminus1 d6( 1113857⟶ minus 1

σminus1 d6( 1113857 σminus1 d7( 1113857⟶ minus 1

σminus1 d7( 1113857 σminus1 d8( 1113857⟶ minus 1

σminus1 d8( 1113857 σminus1 d9( 1113857⟶ minus 1

σminus1 d9( 1113857 σminus1 d10( 1113857⟶ minus 1

σminus1 d10( 1113857 σ0 d11( 1113857⟶ minus 1

σ0 d11( 1113857 σminus1 d12( 1113857⟶ minus 1

σminus1 d12( 1113857 σ0 d13( 1113857⟶ minus 1

σ0 d13( 1113857 σminus1 d14( 1113857⟶ minus 1

σminus1 d14( 1113857 σ2 d15( 1113857⟶ minus 1

σ2 d15( 1113857 σminus1 d16( 1113857⟶ minus 1

σminus1 d16( 1113857 σminus1 d17( 1113857⟶ minus 1

σminus1 d17( 1113857 σminus1 d18( 1113857⟶ minus 1

σminus1 d18( 1113857 σ0 d19( 1113857⟶ minus 1

σ0 d19( 1113857 σminus1 d20( 1113857⟶ minus 1

σminus1 d20( 1113857 σminus1 d21( 1113857⟶ minus 1

σminus1 d21( 1113857 σ0 d22( 1113857⟶ minus 1

σ0 d22( 1113857 σminus1 d23( 1113857⟶ minus 1

σminus1 d23( 1113857 σminus1 d24( 1113857⟶ minus 1

σminus1 d24( 1113857 σminus1 d25( 1113857⟶ minus 1

σminus1 d25( 1113857 σminus1 d26( 1113857⟶ minus 1

σminus1 d26( 1113857 σ0 d27( 1113857⟶ minus 1

σ0 d27( 1113857 σminus1 d28( 1113857⟶ minus 1

σminus1 d28( 1113857 σminus1 d29( 1113857⟶ minus 1

(18)

Table 11 Irregularity of the cryptography objectivesrsquo parameters in the data matrix A for α5

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d1 and d6 qi 3 2 and 1 433 3 072d3 qi 2 and 1 133 3 026d4 qi 1 033 3 006

11117

1 1 1

09095

1105

11115

12

com

plex

ity

alpha cutα=2 α=3 α=5 α=15 α=25

Figure 8 System complexity of the cryptography algorithm designobjectives for different alpha-cuts

12 Computational Intelligence and Neuroscience

410 Calculating the Structure Vectors erefore the firststructure vector based on the software output is

Q 1 1 2( 1113857 (19)

e second structure vector P is

P 1 1 29( 1113857 (20)

Since the large values of P of higher dimensions dem-onstrate more links the second structure vector calculatedfor the alpha of interest shows that the relationship betweenthe cryptography algorithm implementation techniques isminimum

411 Calculating the Obstruction or Inflexibility Vectore obstruction vector (Qlowast) for a sample with α3 24 is

Table 12 Ranking the objectivesrsquo parameters considering different cuts in Q-analysis

e equivalence class for the parameters with minimum relation level (q 0) αSecurity α50 244Speed α30 244Resources α15 122Flexibility α5 40Simplicity-scalability α2 16(All parameters)

securityspeed

optimal usage of resources

flexibility

simplicity-scalability

Figure 9 Prioritizing the design and implementation objectives of the cryptography algorithms using Q-analysis

securityspeed

optimal usage of resources

flexibility

simplicity-scalability-reliability

Figure 10 Prioritizing the design and implementation objectives of the cryptography algorithms using Q-analysis

Table 13 Set of dis and cis the techniques employed for implementing cryptography algorithms

Flexibility C4 Fuzzy logic d25 Storage space d17 Cycle d9 Avalanche effect d1Scalability C5 Software d26 Clustering d18 Multiple step d10 Digital signature d2Speed C6 Steganography d27 Key d19 Hybrid method d11 Block size d3Reliability C7 Music harmony d28 Graph d20 Hardware d12 Image sharing d4

Artificial intelligence d29 Characteristic oriented d21 Hardware d13 Parallel processing d5Security C1 Energy consumption d22 Occupation area d14 reshold technique d6Simplicity C2 Bandwidth consumption d23 Basic science d15 Data mining d7Resources C3 Memory consumption d24 Compression d16 Binary tree d8

Table 14 e incidence matrix of the cryptography algorithmimplementation techniques with α 24

C1 C2 C3 C4 C5 C6 C7

d1 1 0 0 0 0 0 0d2 0 0 0 0 0 0 0d3 0 0 0 0 0 0 0d4 0 0 0 0 0 0 0d5 0 0 0 0 0 0 0d6 0 0 0 0 0 0 0d7 0 0 0 0 0 0 0d8 0 0 0 0 0 0 0d9 0 0 0 0 0 0 0d10 0 0 0 0 0 0 0d11 1 0 0 0 0 0 0d12 0 0 0 0 0 0 0

Computational Intelligence and Neuroscience 13

Qlowast

Q minus I⟶ Qlowast

1 1 21113858 1113859 minus 1 1 11113858 1113859⟶ Qlowast

0 0 11113858 1113859

(21)

e value of QlowastK represents the number of structurallimitations or obstructions for the cryptography techniquesrsquointeraction at dimension k As can be seen from the cal-culated obstruction vector there is a significant relationshipat some levels

412 Calculating Irregularity Irregularity is the integrationdegree of a cryptography method in the total complex Mea-suring irregularity (eccrsquo) for α3 24 is shown in Table 15 Ascan be seen irregularity for the discriminant class d15 indicatesthat this technique is isolated from other techniques

413 Calculating Complexity Structure complexity for asample with α3 24 is

Q 1 1 2( 1113857

ψ(K) 2(2 + 2 + 3)

(3lowast 4)1113890 1113891 117

(22)

e system complexity for different values of alpha isshown in Figure 12 As can be seen the total complexity ofthe system for large values alpha tends to stability

414 Ranking the Cryptography Algorithm ImplementationTechniques ese techniques can be classified into 5 levelsusing the analysis of the obtained results ese levels can beseen in Table 16 and Figure 13 e link power of the factorsin one group is specified with alpha Each level describespriority and importance of the group in the development ofcryptography algorithms Accordingly six methods of basicscience key management using hardware methods usingsteganography Avalanche effect and using hybrid method

q Q21o

112

SETSX15X15X1X11X15X19X27X13 amp X22

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-10-1-1-10-1-1-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-10-1-1-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

2-10-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1

-1-1-1-1-1-1

-1-1-1-1-1

-1-1-1-1

-1-1-1

-1-10

Figure 11 Results of executing the model for cryptography algorithm implementation techniques with α3 24

14 Computational Intelligence and Neuroscience

are the most important cryptography algorithm imple-mentation techniques to achieve the main objectives

5 Conclusion

In this study the design and implementation model of thecryptography algorithms is designed in three levels and itseffective components are extracted To organize the com-ponents in addition to elimination of the additional com-ponents through the measure reduction algorithm a properclassification is applied to examine the mutual effects Afterclassification three 13lowast 4 7lowast 4 and 29lowast 7 matrices areconstituted and the model is implemented on these threematrices To implement the designed model Q-analysis isused Accordingly for support indices five indices of highpriority include human resources research and develop-ment management organizationstructure and equipmentFor the algorithm design objectives index five high priorityindices include security speed optimal usage of resources

and simplicity For the indices related to implementationtechniques of the cryptography algorithms the most appliedtechniques for achieving the determined objectives includeusing basic science hardware and software methods usingkeymanagement hybrid method steganography Avalancheeffect and using artificial intelligence In the future work wewill consider the following cases

A plan should be formulated according to the prioritiesAccording to the outputs of this study and the presented pri-orities the following topics can be investigated in future studies

(1) Presenting a comprehensive model for generatingvarious cryptography algorithms based on the pri-orities of interest

(2) Examining and formulating a model about the roleof basic science for design and implementation ofcryptography algorithms considering the significantrole of ldquobasic sciencerdquo in implementation of thecryptography algorithms and about the role ofschools and universities

(3) Presenting a model for design and implementationof cryptography algorithms in the IoT infrastructurewith optimal resource usage

Data Availability

e data used to support the findings of the study areavailable from the corresponding author upon request

Conflicts of Interest

e authors declare that there are no conflicts of interestregarding the publication of this article

Table 15 Irregularity of the cryptography algorithm implementation techniques for α3 24

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d15 qi 2 1 3 2 1- qi 1 1 2 033

11117

1 1 1

091

1112

com

plex

ity

alpha-cutα=2 α=3 α=5 α=15 α=25

Figure 12 System complexity of cryptography algorithm implementation techniques for different alpha-cuts

Table 16 Ranking the implementation techniques of the cryptography algorithms considering various alpha-cuts using Q-analysis

Equivalence classes for the parameters with minimum relationship level (q 0) αBasic sciences α20 163Key α10 81Hardware and steganography α5 40Avalanche effect and hybrid method α3 24Artificial intelligence software occupation area energy consumption memory consumption α2 16

basic sciences

key

Avalanche effect hybrid method

hardware steganography

artificial intelligence soware occupationarea energy consumption memory consumption

Figure 13 Prioritizing the implementation techniques of thecryptography algorithms using Q-analysis

Computational Intelligence and Neuroscience 15

Acknowledgments

is work was partially supported by Iran National ScienceFoundation (INSF) under contract No 9653979

References

[1] D Nilesh and M Nagle ldquoe new cryptography algorithmwith high throughputrdquo in Proceedings of the 2014 Interna-tional Conference on Computer Communication and Infor-matics pp 3ndash7 Coimbatore India January 2014

[2] S Ravi P Kocher R Lee G McGraw and A RaghunathanldquoSecurity as a new dimension in embedded system designrdquo inProceedings of the 41st annual conference on Design auto-mation-DAC lsquo04 pp 753ndash760 San Diego CA USA March2004

[3] P Kuppuswamy and S Q Y A Khalidi ldquoHybrid encryptiondecryption technique using new public key and symmetric keyalgorithmrdquo International Journal of Information and Com-puter Security vol 6 no 4 pp 372ndash382 2014

[4] S Bhat and V Kapoor ldquoSecure and efficient data privacyauthentication and integrity schemes using hybrid cryptog-raphyrdquo International Conference on Advanced ComputingNetworking and Informatics vol 870 pp 279ndash285 2019

[5] D Ott and C Peikert ldquoAnd other workshop participantsldquoidentifying research challenges in post quantum cryptog-raphy migration and cryptographic agilityrdquo 2019 httparxivorgabs190907353

[6] P Manickam K Shankar E Perumal M Ilayaraja andK Sathesh Kumar ldquoSecure data transmission through reliablevehicles in VANET using optimal lightweight cryptographyrdquoin Advanced Sciences and Technologies for Security Applica-tions pp 193ndash204 Springer New York NY USA 2019

[7] S Prakash and A Rajput ldquoHybrid cryptography for securedata communication in wireless sensor networksrdquo Advancesin Intelligent Systems and Computing vol 696 pp 589ndash5992018

[8] A A Soofi I Riaz and U Rasheed ldquoAn enhanced vigenerecipher for data securityrdquo International Journal of Scientific ampTechnology Research vol 4 no 8 pp 141ndash145 2015

[9] N Shamsuddin M Syafiqah and S Ali Pitchay ldquoImple-menting location-based cryptography on mobile applicationdesign to secure data in cloud storagerdquo Journal of PhysicsConference Series IOP Publishing vol 1551 no 1 2020

[10] L Jiao H Yonglin and D Feng ldquoStream cipher designs areviewrdquo Science China Information Sciences vol 63 no 3pp 1ndash25 2020

[11] NIST ldquoNIST cryptographic standards and guidelines devel-opment processrdquo Inside NIST vol 27 2016

[12] G Alagic J Alperin-Sheriff D Apon et al Status Report onthe First Round of the NIST Post-Quantum CryptographyStandardization Process National Institute of Standards andTechnology Gaithersburg MD USA pp 1ndash27 2019

[13] I Damaj and S Kasbah ldquoAn analysis framework for hardwareand software implementations with applications from cryp-tographyrdquo Computers amp Electrical Engineering vol 69pp 572ndash584 2018

[14] M S Croock Z A Hassan and S D Khudhur ldquoAdaptive keygeneration algorithm based on software engineering meth-odologyrdquo International Journal of Electrical and ComputerEngineering vol 11 no 1 p 589 2021

[15] K Sha T A Yang W Wei and S Davari ldquoA survey of edgecomputing-based designs for iot securityrdquo Digital Commu-nications and Networks vol 6 no 2 pp 195ndash202 2020

[16] M Yahuza M Y I B Idris A W B A Wahab et alldquoSystematic review on security and privacy requirements inedge computing state of the art and future research oppor-tunitiesrdquo IEEE Access vol 8 pp 76541ndash76567 2020

[17] A Bhowmik J Dey and S Karforma A Novel Double TierCryptographic System (nDTCS) to Reinforce Patientsrsquo Privacyin Contemporary COVID-19 Telemedicine Springer NewYork NY USA 2021

[18] N Gupta User Integrity Protection Security Model forEnhanced Telemedicine for Healthcare Networks 2020httpsosfio69jdq

[19] J Dey A Bhowmik A Sarkar S Karforma andB Chowdhury Cryptographic Engineering on COVID-19Telemedicine An Intelligent Transmission through RecurrentRelation Based Session Key Springer New York NY USA2021

[20] V Hosseinian H Ayatollahi H Haghani and E MehraeenldquoRequirements of information security in a telemedicinenetwork review of IT managersrsquo opinionrdquo Journal of Para-medical Sciences amp Rehabilitation vol 4 no 2 pp 31ndash402015

[21] D Karaoglan Altop A Levi and V Tuzcu ldquoDeriving cryp-tographic keys from physiological signalsrdquo Pervasive andMobile Computing vol 39 pp 65ndash79 2017

[22] O Uzunkol and M S Kiraz ldquoStill wrong use of pairings incryptographyrdquo Applied Mathematics and Computationvol 333 pp 467ndash479 2018

[23] S Picek D Jakobovic J F Miller L Batina and M CupicldquoCryptographic Boolean functions one output many designcriteriardquo Applied Soft Computing vol 40 pp 635ndash653 2016

[24] R Saha and G Geetha ldquoSymmetric random function gen-erator (SRFG) a novel cryptographic primitive for designingfast and robust algorithmsrdquo Chaos Solitons amp Fractalsvol 104 pp 371ndash377 2017

[25] A Achuthshankar and A Achuthshankar ldquoA novel sym-metric cryptography algorithm for fast and secure encryp-tionrdquo in Proceedings of the 2015 IEEE 9th InternationalConference on Intelligent Systems and Control (ISCO) pp 4ndash9Coimbatore India January 2015

[26] D Ott and C Peikert ldquoIdentifying research challenges in postquantum cryptography migration and cryptographic agilityrdquo2019 httparxivorgabs190907353

[27] W Liu B Ying H Yang and H Wang ldquoAccurate modelingfor predicting cryptography overheads on wireless sensornodesrdquovol 2 pp 997ndash1001 in Proceedings of the 2019 11thInternational Conference on Advanced CommunicationTechnology vol 2 pp 997ndash1001 IEEE Bangalore IndiaFebruary 2019

[28] S Feizi A Ahmadi and A Nemati ldquoA hardware imple-mentation of simon cryptography algorithmrdquo in Proceedingsof the 2014 4th International Conference on Computer andKnowledge Engineering (ICCKE) pp 245ndash250 IEEEMashhad Iran 2014

[29] E ambiraja G Ramesh and R Umarani ldquoA survey onvarious most common encryption techniquesrdquo 2012 httpwwwijarcssecom

[30] A Gupta and N K Walia ldquoCryptography Algorithms areviewrdquo International Journal of Engineering Research andDevelopment vol 2 no 2 pp 1667ndash1672 2014

[31] R Bhanot and R Hans ldquoA review and comparative analysis ofvarious encryption algorithmsrdquo International Journal of Se-curity and Its Applications vol 9 no 4 pp 289ndash306 2015

[32] S Chandra S Paira S S Alam and G Sanyal ldquoA com-parative survey of symmetric and asymmetric key

16 Computational Intelligence and Neuroscience

cryptographyrdquo in Proceedings of the 2014 InternationalConference on Electronics Communication and Computa-tional Engineering (ICECCE) pp 83ndash93 Hosur India No-vember 2014

[33] F Maqsood M Ahmed M Mumtaz and M Ali ldquoCryp-tography a comparative analysis for modern techniquesrdquoInternational Journal of Advanced Computer Science andApplications vol 8 no 6 p 6 2017

[34] S B Sadkhan and A O Salman ldquoA survey on lightweight-cryptography status and future challengesrdquo in Proceedings ofthe International Conference on Advances in SustainableEngineering and Applications pp 105ndash108 Wasit Iraq June2018

[35] N V oai ldquoCriteria and dimension reduction of linearmultiple criteria optimization problemsrdquo Journal of GlobalOptimization vol 52 no 3 pp 499ndash508 2012

[36] C E Shannon ldquoCommunication theory of secrecy systems1945rdquo MD Computing vol 15 no 1 pp 57ndash64 1998

[37] L J Fennelly M Beaudry and M A Perry Security in 2025[38] ITU Global Cybersecurity Index 2018 2019[39] ITU Global Cybersecurity Index (GCI) 2017 2017

Computational Intelligence and Neuroscience 17

Qlowast

Q minus I⟶ Qlowast

2 1 1 31113858 1113859 minus 1 1 1 11113858 1113859⟶ Qlowast

0 0 0 01113858 1113859

(13)

According to the obtained values it can be concludedthat there are no structural obstructions among the main

design indices of the cryptography algorithms at ismultiple objectives are considered simultaneously by thecryptography algorithm designers

Table 7 e irregularity of the data matrix parameters for α= 7

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d2 d5 d6 d8 qi 3 2 1 364 3 061d9 d10d11 qi 2 1 064 3 011d4 d7 d12 qi 1 014 3 002

human resourceRampD

rules FAVA standard

management infrastructure

organizationstructure equipmentinternational and public relations finanacial

resources and material

Figure 6 e ranking pyramid of the support components using Q-analysis

Table 8 Ranking of the support components using Q-analysis

Relationship of the support components with the design and implementation of the cryptography algorithms (q 0)No relationship α 0 complete relationship α 100

Human resource RampD rules FAVA and standard α100 10Management and infrastructure α90 9Organizationstructure and equipment α80 8International and public relations financial resources and material α70 7Culture α60 6All components α60 5

Table 9 Set of dis and cis the objectivesrsquo indices

eoretical basis C2 Reliability d7 Scalability d5 Resources d3 Security d1Implementation C3 Application C1 Speed d6 Flexibility d4 Simplicity d2Evaluation C4

Table 10 e incidence matrix of the design objectives of thecryptography algorithms with α= 40

c1 c2 c3 c4

d1 1 1 1 1d2 0 0 0 0d3 1 0 1 1d4 0 0 1 1d5 0 0 0 0d6 1 1 1 1d7 0 0 0 0

-1 -1 -1 -1 -1

2 1 -1 2

1 -1 1

-1 -1

3

3 -1 2 1 -1 3 -1

-1

-1

-1

-1

-1

-1q Q3210

11

11

SETSX1X6X1X3X6X1X3X4X6X1X3X4X6

Figure 7 e results of executing the model for design objectivesrsquoindices with α 40

10 Computational Intelligence and Neuroscience

43 Calculating Irregularity e results of applying theChinese method (ecclsquo(σ)) for calculating irregularity forα5 40 are shown in Table 11 e calculated irregu-larity value shows that indices d1 (security) and d6(speed) have received more attention compared to otherindices

44 Calculating Complexity Structural complexity for asample with α5 40 is

dimensions 3 2 1 0

Q Qdim3 Qdim2 Qdim1 Qdim0( 1113857

Q 1 1 1 1( 1113857

ψ(K) 2(1 + 2 + 3 + 4)

(4lowast 5)1113890 1113891 1

(14)

Since in the above model there was no obstructionamong the components of the equivalent classes in any of thecommunication levels it was expected that there is not ahigh complexity among objectives of the cryptography al-gorithm design and the complexity index of 1 verified thisexpectation e system complexity for different alpha-cutsis shown in Figure 8 As can be seen the system complexity is1 for all significant values of alpha

45 Prioritizing the Parameters of the Main Objectives of theCryptography Algorithm Design Using Q-AnalysisAccording to the analysis results the parameters can beclassified into 5 levelsese levels are shown in Table 12 andFigure 9 Each level indicates a priority and importance ofthe group in the development of cryptography algorithmsTo allocate proper resources the components at the higherlevels of the pyramid are of higher priority As can be seenthree objectives of security speed and optimal usage ofresources have the highest priority for the design of cryp-tography algorithms

46 Executing the Model and Analyzing the Results for theCryptography Algorithm Implementation Techniques einteraction matrix between the 29 extracted techniques andthe sevenmain objectives is shown in Figure 10e purposeof this section is to rank the cryptography algorithmimplementation techniques to achieve the goals of interest

47Calculating the IncidenceMatrixandExecuting theModele data matrix A is comprised of two sets e set Drepresents the employed techniques and the set C representsthe seven main design objectives (Table 13)

D d1 d2 d291113864 1113865

C c1 c2 c71113864 1113865(15)

Some parts of the incidence matrix calculated from thedata matrix A for α3 are shown in Table 14 and the resultsof theQ-analysis model for α3 24 are shown in Figure 11

48 Geometric Representation In the sample with α3 24the dis are

d1 c11113864 1113865

d2

d3

d4

d5

d6

d7

d8

d9

d10

d11 c11113864 1113865

d12

d13 c51113864 1113865

d14

d15 c1 c4 c51113864 1113865

d16

d17

d18

d19 c11113864 1113865

d20

d21

d22 c31113864 1113865

d23

d24

d25

d26

d27 c11113864 1113865

d28

d29

(16)

e simplexes of σp(di) are

Computational Intelligence and Neuroscience 11

σ0 d1( 1113857

σminus1 d2( 1113857

σminus1 d3( 1113857

σminus1 d4( 1113857

σminus1 d5( 1113857

σminus1 d6( 1113857

σminus1 d7( 1113857

σminus1 d8( 1113857

σminus1 d9( 1113857

σminus1 d10( 1113857

σ0 d11( 1113857

σminus1 d12( 1113857

σ0 d13( 1113857

σminus1 d14( 1113857

σ2 d15( 1113857

σminus1 d16( 1113857

σminus1 d17( 1113857

σminus1 d18( 1113857

σ0 d19( 1113857

σminus1 d20( 1113857

σminus1 d21( 1113857

σ0 d22( 1113857

σminus1 d23( 1113857

σminus1 d24( 1113857

σminus1 d25( 1113857

σminus1 d26( 1113857

σ0 d27( 1113857

σminus1 d28( 1113857

σminus1 d29( 1113857

(17)

erefore the complex dimension is 2 In other wordsthe discriminant class d15 (basic sciences) has the largestdimension

49 Calculating the Dimensions and the Q-Link In the alphadefined for cryptography algorithm implementation tech-niques to achieve the defined goals the obtained q-linkshows that these techniques are relatively independent al-though there is a weak relationship between some tech-niques Q-link in the samples with α3 24 between eachtwo subsequent di is as follows

σ0 d1( 1113857 σminus1 d2( 1113857⟶ minus 1

σminus1 d2( 1113857 σminus1 d3( 1113857⟶ minus 1

σ0 d3( 1113857 σminus1 d4( 1113857⟶ minus 1

σminus1 d4( 1113857 σminus1 d5( 1113857⟶ minus 1

σminus1 d5( 1113857 σminus1 d6( 1113857⟶ minus 1

σminus1 d6( 1113857 σminus1 d7( 1113857⟶ minus 1

σminus1 d7( 1113857 σminus1 d8( 1113857⟶ minus 1

σminus1 d8( 1113857 σminus1 d9( 1113857⟶ minus 1

σminus1 d9( 1113857 σminus1 d10( 1113857⟶ minus 1

σminus1 d10( 1113857 σ0 d11( 1113857⟶ minus 1

σ0 d11( 1113857 σminus1 d12( 1113857⟶ minus 1

σminus1 d12( 1113857 σ0 d13( 1113857⟶ minus 1

σ0 d13( 1113857 σminus1 d14( 1113857⟶ minus 1

σminus1 d14( 1113857 σ2 d15( 1113857⟶ minus 1

σ2 d15( 1113857 σminus1 d16( 1113857⟶ minus 1

σminus1 d16( 1113857 σminus1 d17( 1113857⟶ minus 1

σminus1 d17( 1113857 σminus1 d18( 1113857⟶ minus 1

σminus1 d18( 1113857 σ0 d19( 1113857⟶ minus 1

σ0 d19( 1113857 σminus1 d20( 1113857⟶ minus 1

σminus1 d20( 1113857 σminus1 d21( 1113857⟶ minus 1

σminus1 d21( 1113857 σ0 d22( 1113857⟶ minus 1

σ0 d22( 1113857 σminus1 d23( 1113857⟶ minus 1

σminus1 d23( 1113857 σminus1 d24( 1113857⟶ minus 1

σminus1 d24( 1113857 σminus1 d25( 1113857⟶ minus 1

σminus1 d25( 1113857 σminus1 d26( 1113857⟶ minus 1

σminus1 d26( 1113857 σ0 d27( 1113857⟶ minus 1

σ0 d27( 1113857 σminus1 d28( 1113857⟶ minus 1

σminus1 d28( 1113857 σminus1 d29( 1113857⟶ minus 1

(18)

Table 11 Irregularity of the cryptography objectivesrsquo parameters in the data matrix A for α5

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d1 and d6 qi 3 2 and 1 433 3 072d3 qi 2 and 1 133 3 026d4 qi 1 033 3 006

11117

1 1 1

09095

1105

11115

12

com

plex

ity

alpha cutα=2 α=3 α=5 α=15 α=25

Figure 8 System complexity of the cryptography algorithm designobjectives for different alpha-cuts

12 Computational Intelligence and Neuroscience

410 Calculating the Structure Vectors erefore the firststructure vector based on the software output is

Q 1 1 2( 1113857 (19)

e second structure vector P is

P 1 1 29( 1113857 (20)

Since the large values of P of higher dimensions dem-onstrate more links the second structure vector calculatedfor the alpha of interest shows that the relationship betweenthe cryptography algorithm implementation techniques isminimum

411 Calculating the Obstruction or Inflexibility Vectore obstruction vector (Qlowast) for a sample with α3 24 is

Table 12 Ranking the objectivesrsquo parameters considering different cuts in Q-analysis

e equivalence class for the parameters with minimum relation level (q 0) αSecurity α50 244Speed α30 244Resources α15 122Flexibility α5 40Simplicity-scalability α2 16(All parameters)

securityspeed

optimal usage of resources

flexibility

simplicity-scalability

Figure 9 Prioritizing the design and implementation objectives of the cryptography algorithms using Q-analysis

securityspeed

optimal usage of resources

flexibility

simplicity-scalability-reliability

Figure 10 Prioritizing the design and implementation objectives of the cryptography algorithms using Q-analysis

Table 13 Set of dis and cis the techniques employed for implementing cryptography algorithms

Flexibility C4 Fuzzy logic d25 Storage space d17 Cycle d9 Avalanche effect d1Scalability C5 Software d26 Clustering d18 Multiple step d10 Digital signature d2Speed C6 Steganography d27 Key d19 Hybrid method d11 Block size d3Reliability C7 Music harmony d28 Graph d20 Hardware d12 Image sharing d4

Artificial intelligence d29 Characteristic oriented d21 Hardware d13 Parallel processing d5Security C1 Energy consumption d22 Occupation area d14 reshold technique d6Simplicity C2 Bandwidth consumption d23 Basic science d15 Data mining d7Resources C3 Memory consumption d24 Compression d16 Binary tree d8

Table 14 e incidence matrix of the cryptography algorithmimplementation techniques with α 24

C1 C2 C3 C4 C5 C6 C7

d1 1 0 0 0 0 0 0d2 0 0 0 0 0 0 0d3 0 0 0 0 0 0 0d4 0 0 0 0 0 0 0d5 0 0 0 0 0 0 0d6 0 0 0 0 0 0 0d7 0 0 0 0 0 0 0d8 0 0 0 0 0 0 0d9 0 0 0 0 0 0 0d10 0 0 0 0 0 0 0d11 1 0 0 0 0 0 0d12 0 0 0 0 0 0 0

Computational Intelligence and Neuroscience 13

Qlowast

Q minus I⟶ Qlowast

1 1 21113858 1113859 minus 1 1 11113858 1113859⟶ Qlowast

0 0 11113858 1113859

(21)

e value of QlowastK represents the number of structurallimitations or obstructions for the cryptography techniquesrsquointeraction at dimension k As can be seen from the cal-culated obstruction vector there is a significant relationshipat some levels

412 Calculating Irregularity Irregularity is the integrationdegree of a cryptography method in the total complex Mea-suring irregularity (eccrsquo) for α3 24 is shown in Table 15 Ascan be seen irregularity for the discriminant class d15 indicatesthat this technique is isolated from other techniques

413 Calculating Complexity Structure complexity for asample with α3 24 is

Q 1 1 2( 1113857

ψ(K) 2(2 + 2 + 3)

(3lowast 4)1113890 1113891 117

(22)

e system complexity for different values of alpha isshown in Figure 12 As can be seen the total complexity ofthe system for large values alpha tends to stability

414 Ranking the Cryptography Algorithm ImplementationTechniques ese techniques can be classified into 5 levelsusing the analysis of the obtained results ese levels can beseen in Table 16 and Figure 13 e link power of the factorsin one group is specified with alpha Each level describespriority and importance of the group in the development ofcryptography algorithms Accordingly six methods of basicscience key management using hardware methods usingsteganography Avalanche effect and using hybrid method

q Q21o

112

SETSX15X15X1X11X15X19X27X13 amp X22

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-10-1-1-10-1-1-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-10-1-1-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

2-10-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1

-1-1-1-1-1-1

-1-1-1-1-1

-1-1-1-1

-1-1-1

-1-10

Figure 11 Results of executing the model for cryptography algorithm implementation techniques with α3 24

14 Computational Intelligence and Neuroscience

are the most important cryptography algorithm imple-mentation techniques to achieve the main objectives

5 Conclusion

In this study the design and implementation model of thecryptography algorithms is designed in three levels and itseffective components are extracted To organize the com-ponents in addition to elimination of the additional com-ponents through the measure reduction algorithm a properclassification is applied to examine the mutual effects Afterclassification three 13lowast 4 7lowast 4 and 29lowast 7 matrices areconstituted and the model is implemented on these threematrices To implement the designed model Q-analysis isused Accordingly for support indices five indices of highpriority include human resources research and develop-ment management organizationstructure and equipmentFor the algorithm design objectives index five high priorityindices include security speed optimal usage of resources

and simplicity For the indices related to implementationtechniques of the cryptography algorithms the most appliedtechniques for achieving the determined objectives includeusing basic science hardware and software methods usingkeymanagement hybrid method steganography Avalancheeffect and using artificial intelligence In the future work wewill consider the following cases

A plan should be formulated according to the prioritiesAccording to the outputs of this study and the presented pri-orities the following topics can be investigated in future studies

(1) Presenting a comprehensive model for generatingvarious cryptography algorithms based on the pri-orities of interest

(2) Examining and formulating a model about the roleof basic science for design and implementation ofcryptography algorithms considering the significantrole of ldquobasic sciencerdquo in implementation of thecryptography algorithms and about the role ofschools and universities

(3) Presenting a model for design and implementationof cryptography algorithms in the IoT infrastructurewith optimal resource usage

Data Availability

e data used to support the findings of the study areavailable from the corresponding author upon request

Conflicts of Interest

e authors declare that there are no conflicts of interestregarding the publication of this article

Table 15 Irregularity of the cryptography algorithm implementation techniques for α3 24

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d15 qi 2 1 3 2 1- qi 1 1 2 033

11117

1 1 1

091

1112

com

plex

ity

alpha-cutα=2 α=3 α=5 α=15 α=25

Figure 12 System complexity of cryptography algorithm implementation techniques for different alpha-cuts

Table 16 Ranking the implementation techniques of the cryptography algorithms considering various alpha-cuts using Q-analysis

Equivalence classes for the parameters with minimum relationship level (q 0) αBasic sciences α20 163Key α10 81Hardware and steganography α5 40Avalanche effect and hybrid method α3 24Artificial intelligence software occupation area energy consumption memory consumption α2 16

basic sciences

key

Avalanche effect hybrid method

hardware steganography

artificial intelligence soware occupationarea energy consumption memory consumption

Figure 13 Prioritizing the implementation techniques of thecryptography algorithms using Q-analysis

Computational Intelligence and Neuroscience 15

Acknowledgments

is work was partially supported by Iran National ScienceFoundation (INSF) under contract No 9653979

References

[1] D Nilesh and M Nagle ldquoe new cryptography algorithmwith high throughputrdquo in Proceedings of the 2014 Interna-tional Conference on Computer Communication and Infor-matics pp 3ndash7 Coimbatore India January 2014

[2] S Ravi P Kocher R Lee G McGraw and A RaghunathanldquoSecurity as a new dimension in embedded system designrdquo inProceedings of the 41st annual conference on Design auto-mation-DAC lsquo04 pp 753ndash760 San Diego CA USA March2004

[3] P Kuppuswamy and S Q Y A Khalidi ldquoHybrid encryptiondecryption technique using new public key and symmetric keyalgorithmrdquo International Journal of Information and Com-puter Security vol 6 no 4 pp 372ndash382 2014

[4] S Bhat and V Kapoor ldquoSecure and efficient data privacyauthentication and integrity schemes using hybrid cryptog-raphyrdquo International Conference on Advanced ComputingNetworking and Informatics vol 870 pp 279ndash285 2019

[5] D Ott and C Peikert ldquoAnd other workshop participantsldquoidentifying research challenges in post quantum cryptog-raphy migration and cryptographic agilityrdquo 2019 httparxivorgabs190907353

[6] P Manickam K Shankar E Perumal M Ilayaraja andK Sathesh Kumar ldquoSecure data transmission through reliablevehicles in VANET using optimal lightweight cryptographyrdquoin Advanced Sciences and Technologies for Security Applica-tions pp 193ndash204 Springer New York NY USA 2019

[7] S Prakash and A Rajput ldquoHybrid cryptography for securedata communication in wireless sensor networksrdquo Advancesin Intelligent Systems and Computing vol 696 pp 589ndash5992018

[8] A A Soofi I Riaz and U Rasheed ldquoAn enhanced vigenerecipher for data securityrdquo International Journal of Scientific ampTechnology Research vol 4 no 8 pp 141ndash145 2015

[9] N Shamsuddin M Syafiqah and S Ali Pitchay ldquoImple-menting location-based cryptography on mobile applicationdesign to secure data in cloud storagerdquo Journal of PhysicsConference Series IOP Publishing vol 1551 no 1 2020

[10] L Jiao H Yonglin and D Feng ldquoStream cipher designs areviewrdquo Science China Information Sciences vol 63 no 3pp 1ndash25 2020

[11] NIST ldquoNIST cryptographic standards and guidelines devel-opment processrdquo Inside NIST vol 27 2016

[12] G Alagic J Alperin-Sheriff D Apon et al Status Report onthe First Round of the NIST Post-Quantum CryptographyStandardization Process National Institute of Standards andTechnology Gaithersburg MD USA pp 1ndash27 2019

[13] I Damaj and S Kasbah ldquoAn analysis framework for hardwareand software implementations with applications from cryp-tographyrdquo Computers amp Electrical Engineering vol 69pp 572ndash584 2018

[14] M S Croock Z A Hassan and S D Khudhur ldquoAdaptive keygeneration algorithm based on software engineering meth-odologyrdquo International Journal of Electrical and ComputerEngineering vol 11 no 1 p 589 2021

[15] K Sha T A Yang W Wei and S Davari ldquoA survey of edgecomputing-based designs for iot securityrdquo Digital Commu-nications and Networks vol 6 no 2 pp 195ndash202 2020

[16] M Yahuza M Y I B Idris A W B A Wahab et alldquoSystematic review on security and privacy requirements inedge computing state of the art and future research oppor-tunitiesrdquo IEEE Access vol 8 pp 76541ndash76567 2020

[17] A Bhowmik J Dey and S Karforma A Novel Double TierCryptographic System (nDTCS) to Reinforce Patientsrsquo Privacyin Contemporary COVID-19 Telemedicine Springer NewYork NY USA 2021

[18] N Gupta User Integrity Protection Security Model forEnhanced Telemedicine for Healthcare Networks 2020httpsosfio69jdq

[19] J Dey A Bhowmik A Sarkar S Karforma andB Chowdhury Cryptographic Engineering on COVID-19Telemedicine An Intelligent Transmission through RecurrentRelation Based Session Key Springer New York NY USA2021

[20] V Hosseinian H Ayatollahi H Haghani and E MehraeenldquoRequirements of information security in a telemedicinenetwork review of IT managersrsquo opinionrdquo Journal of Para-medical Sciences amp Rehabilitation vol 4 no 2 pp 31ndash402015

[21] D Karaoglan Altop A Levi and V Tuzcu ldquoDeriving cryp-tographic keys from physiological signalsrdquo Pervasive andMobile Computing vol 39 pp 65ndash79 2017

[22] O Uzunkol and M S Kiraz ldquoStill wrong use of pairings incryptographyrdquo Applied Mathematics and Computationvol 333 pp 467ndash479 2018

[23] S Picek D Jakobovic J F Miller L Batina and M CupicldquoCryptographic Boolean functions one output many designcriteriardquo Applied Soft Computing vol 40 pp 635ndash653 2016

[24] R Saha and G Geetha ldquoSymmetric random function gen-erator (SRFG) a novel cryptographic primitive for designingfast and robust algorithmsrdquo Chaos Solitons amp Fractalsvol 104 pp 371ndash377 2017

[25] A Achuthshankar and A Achuthshankar ldquoA novel sym-metric cryptography algorithm for fast and secure encryp-tionrdquo in Proceedings of the 2015 IEEE 9th InternationalConference on Intelligent Systems and Control (ISCO) pp 4ndash9Coimbatore India January 2015

[26] D Ott and C Peikert ldquoIdentifying research challenges in postquantum cryptography migration and cryptographic agilityrdquo2019 httparxivorgabs190907353

[27] W Liu B Ying H Yang and H Wang ldquoAccurate modelingfor predicting cryptography overheads on wireless sensornodesrdquovol 2 pp 997ndash1001 in Proceedings of the 2019 11thInternational Conference on Advanced CommunicationTechnology vol 2 pp 997ndash1001 IEEE Bangalore IndiaFebruary 2019

[28] S Feizi A Ahmadi and A Nemati ldquoA hardware imple-mentation of simon cryptography algorithmrdquo in Proceedingsof the 2014 4th International Conference on Computer andKnowledge Engineering (ICCKE) pp 245ndash250 IEEEMashhad Iran 2014

[29] E ambiraja G Ramesh and R Umarani ldquoA survey onvarious most common encryption techniquesrdquo 2012 httpwwwijarcssecom

[30] A Gupta and N K Walia ldquoCryptography Algorithms areviewrdquo International Journal of Engineering Research andDevelopment vol 2 no 2 pp 1667ndash1672 2014

[31] R Bhanot and R Hans ldquoA review and comparative analysis ofvarious encryption algorithmsrdquo International Journal of Se-curity and Its Applications vol 9 no 4 pp 289ndash306 2015

[32] S Chandra S Paira S S Alam and G Sanyal ldquoA com-parative survey of symmetric and asymmetric key

16 Computational Intelligence and Neuroscience

cryptographyrdquo in Proceedings of the 2014 InternationalConference on Electronics Communication and Computa-tional Engineering (ICECCE) pp 83ndash93 Hosur India No-vember 2014

[33] F Maqsood M Ahmed M Mumtaz and M Ali ldquoCryp-tography a comparative analysis for modern techniquesrdquoInternational Journal of Advanced Computer Science andApplications vol 8 no 6 p 6 2017

[34] S B Sadkhan and A O Salman ldquoA survey on lightweight-cryptography status and future challengesrdquo in Proceedings ofthe International Conference on Advances in SustainableEngineering and Applications pp 105ndash108 Wasit Iraq June2018

[35] N V oai ldquoCriteria and dimension reduction of linearmultiple criteria optimization problemsrdquo Journal of GlobalOptimization vol 52 no 3 pp 499ndash508 2012

[36] C E Shannon ldquoCommunication theory of secrecy systems1945rdquo MD Computing vol 15 no 1 pp 57ndash64 1998

[37] L J Fennelly M Beaudry and M A Perry Security in 2025[38] ITU Global Cybersecurity Index 2018 2019[39] ITU Global Cybersecurity Index (GCI) 2017 2017

Computational Intelligence and Neuroscience 17

43 Calculating Irregularity e results of applying theChinese method (ecclsquo(σ)) for calculating irregularity forα5 40 are shown in Table 11 e calculated irregu-larity value shows that indices d1 (security) and d6(speed) have received more attention compared to otherindices

44 Calculating Complexity Structural complexity for asample with α5 40 is

dimensions 3 2 1 0

Q Qdim3 Qdim2 Qdim1 Qdim0( 1113857

Q 1 1 1 1( 1113857

ψ(K) 2(1 + 2 + 3 + 4)

(4lowast 5)1113890 1113891 1

(14)

Since in the above model there was no obstructionamong the components of the equivalent classes in any of thecommunication levels it was expected that there is not ahigh complexity among objectives of the cryptography al-gorithm design and the complexity index of 1 verified thisexpectation e system complexity for different alpha-cutsis shown in Figure 8 As can be seen the system complexity is1 for all significant values of alpha

45 Prioritizing the Parameters of the Main Objectives of theCryptography Algorithm Design Using Q-AnalysisAccording to the analysis results the parameters can beclassified into 5 levelsese levels are shown in Table 12 andFigure 9 Each level indicates a priority and importance ofthe group in the development of cryptography algorithmsTo allocate proper resources the components at the higherlevels of the pyramid are of higher priority As can be seenthree objectives of security speed and optimal usage ofresources have the highest priority for the design of cryp-tography algorithms

46 Executing the Model and Analyzing the Results for theCryptography Algorithm Implementation Techniques einteraction matrix between the 29 extracted techniques andthe sevenmain objectives is shown in Figure 10e purposeof this section is to rank the cryptography algorithmimplementation techniques to achieve the goals of interest

47Calculating the IncidenceMatrixandExecuting theModele data matrix A is comprised of two sets e set Drepresents the employed techniques and the set C representsthe seven main design objectives (Table 13)

D d1 d2 d291113864 1113865

C c1 c2 c71113864 1113865(15)

Some parts of the incidence matrix calculated from thedata matrix A for α3 are shown in Table 14 and the resultsof theQ-analysis model for α3 24 are shown in Figure 11

48 Geometric Representation In the sample with α3 24the dis are

d1 c11113864 1113865

d2

d3

d4

d5

d6

d7

d8

d9

d10

d11 c11113864 1113865

d12

d13 c51113864 1113865

d14

d15 c1 c4 c51113864 1113865

d16

d17

d18

d19 c11113864 1113865

d20

d21

d22 c31113864 1113865

d23

d24

d25

d26

d27 c11113864 1113865

d28

d29

(16)

e simplexes of σp(di) are

Computational Intelligence and Neuroscience 11

σ0 d1( 1113857

σminus1 d2( 1113857

σminus1 d3( 1113857

σminus1 d4( 1113857

σminus1 d5( 1113857

σminus1 d6( 1113857

σminus1 d7( 1113857

σminus1 d8( 1113857

σminus1 d9( 1113857

σminus1 d10( 1113857

σ0 d11( 1113857

σminus1 d12( 1113857

σ0 d13( 1113857

σminus1 d14( 1113857

σ2 d15( 1113857

σminus1 d16( 1113857

σminus1 d17( 1113857

σminus1 d18( 1113857

σ0 d19( 1113857

σminus1 d20( 1113857

σminus1 d21( 1113857

σ0 d22( 1113857

σminus1 d23( 1113857

σminus1 d24( 1113857

σminus1 d25( 1113857

σminus1 d26( 1113857

σ0 d27( 1113857

σminus1 d28( 1113857

σminus1 d29( 1113857

(17)

erefore the complex dimension is 2 In other wordsthe discriminant class d15 (basic sciences) has the largestdimension

49 Calculating the Dimensions and the Q-Link In the alphadefined for cryptography algorithm implementation tech-niques to achieve the defined goals the obtained q-linkshows that these techniques are relatively independent al-though there is a weak relationship between some tech-niques Q-link in the samples with α3 24 between eachtwo subsequent di is as follows

σ0 d1( 1113857 σminus1 d2( 1113857⟶ minus 1

σminus1 d2( 1113857 σminus1 d3( 1113857⟶ minus 1

σ0 d3( 1113857 σminus1 d4( 1113857⟶ minus 1

σminus1 d4( 1113857 σminus1 d5( 1113857⟶ minus 1

σminus1 d5( 1113857 σminus1 d6( 1113857⟶ minus 1

σminus1 d6( 1113857 σminus1 d7( 1113857⟶ minus 1

σminus1 d7( 1113857 σminus1 d8( 1113857⟶ minus 1

σminus1 d8( 1113857 σminus1 d9( 1113857⟶ minus 1

σminus1 d9( 1113857 σminus1 d10( 1113857⟶ minus 1

σminus1 d10( 1113857 σ0 d11( 1113857⟶ minus 1

σ0 d11( 1113857 σminus1 d12( 1113857⟶ minus 1

σminus1 d12( 1113857 σ0 d13( 1113857⟶ minus 1

σ0 d13( 1113857 σminus1 d14( 1113857⟶ minus 1

σminus1 d14( 1113857 σ2 d15( 1113857⟶ minus 1

σ2 d15( 1113857 σminus1 d16( 1113857⟶ minus 1

σminus1 d16( 1113857 σminus1 d17( 1113857⟶ minus 1

σminus1 d17( 1113857 σminus1 d18( 1113857⟶ minus 1

σminus1 d18( 1113857 σ0 d19( 1113857⟶ minus 1

σ0 d19( 1113857 σminus1 d20( 1113857⟶ minus 1

σminus1 d20( 1113857 σminus1 d21( 1113857⟶ minus 1

σminus1 d21( 1113857 σ0 d22( 1113857⟶ minus 1

σ0 d22( 1113857 σminus1 d23( 1113857⟶ minus 1

σminus1 d23( 1113857 σminus1 d24( 1113857⟶ minus 1

σminus1 d24( 1113857 σminus1 d25( 1113857⟶ minus 1

σminus1 d25( 1113857 σminus1 d26( 1113857⟶ minus 1

σminus1 d26( 1113857 σ0 d27( 1113857⟶ minus 1

σ0 d27( 1113857 σminus1 d28( 1113857⟶ minus 1

σminus1 d28( 1113857 σminus1 d29( 1113857⟶ minus 1

(18)

Table 11 Irregularity of the cryptography objectivesrsquo parameters in the data matrix A for α5

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d1 and d6 qi 3 2 and 1 433 3 072d3 qi 2 and 1 133 3 026d4 qi 1 033 3 006

11117

1 1 1

09095

1105

11115

12

com

plex

ity

alpha cutα=2 α=3 α=5 α=15 α=25

Figure 8 System complexity of the cryptography algorithm designobjectives for different alpha-cuts

12 Computational Intelligence and Neuroscience

410 Calculating the Structure Vectors erefore the firststructure vector based on the software output is

Q 1 1 2( 1113857 (19)

e second structure vector P is

P 1 1 29( 1113857 (20)

Since the large values of P of higher dimensions dem-onstrate more links the second structure vector calculatedfor the alpha of interest shows that the relationship betweenthe cryptography algorithm implementation techniques isminimum

411 Calculating the Obstruction or Inflexibility Vectore obstruction vector (Qlowast) for a sample with α3 24 is

Table 12 Ranking the objectivesrsquo parameters considering different cuts in Q-analysis

e equivalence class for the parameters with minimum relation level (q 0) αSecurity α50 244Speed α30 244Resources α15 122Flexibility α5 40Simplicity-scalability α2 16(All parameters)

securityspeed

optimal usage of resources

flexibility

simplicity-scalability

Figure 9 Prioritizing the design and implementation objectives of the cryptography algorithms using Q-analysis

securityspeed

optimal usage of resources

flexibility

simplicity-scalability-reliability

Figure 10 Prioritizing the design and implementation objectives of the cryptography algorithms using Q-analysis

Table 13 Set of dis and cis the techniques employed for implementing cryptography algorithms

Flexibility C4 Fuzzy logic d25 Storage space d17 Cycle d9 Avalanche effect d1Scalability C5 Software d26 Clustering d18 Multiple step d10 Digital signature d2Speed C6 Steganography d27 Key d19 Hybrid method d11 Block size d3Reliability C7 Music harmony d28 Graph d20 Hardware d12 Image sharing d4

Artificial intelligence d29 Characteristic oriented d21 Hardware d13 Parallel processing d5Security C1 Energy consumption d22 Occupation area d14 reshold technique d6Simplicity C2 Bandwidth consumption d23 Basic science d15 Data mining d7Resources C3 Memory consumption d24 Compression d16 Binary tree d8

Table 14 e incidence matrix of the cryptography algorithmimplementation techniques with α 24

C1 C2 C3 C4 C5 C6 C7

d1 1 0 0 0 0 0 0d2 0 0 0 0 0 0 0d3 0 0 0 0 0 0 0d4 0 0 0 0 0 0 0d5 0 0 0 0 0 0 0d6 0 0 0 0 0 0 0d7 0 0 0 0 0 0 0d8 0 0 0 0 0 0 0d9 0 0 0 0 0 0 0d10 0 0 0 0 0 0 0d11 1 0 0 0 0 0 0d12 0 0 0 0 0 0 0

Computational Intelligence and Neuroscience 13

Qlowast

Q minus I⟶ Qlowast

1 1 21113858 1113859 minus 1 1 11113858 1113859⟶ Qlowast

0 0 11113858 1113859

(21)

e value of QlowastK represents the number of structurallimitations or obstructions for the cryptography techniquesrsquointeraction at dimension k As can be seen from the cal-culated obstruction vector there is a significant relationshipat some levels

412 Calculating Irregularity Irregularity is the integrationdegree of a cryptography method in the total complex Mea-suring irregularity (eccrsquo) for α3 24 is shown in Table 15 Ascan be seen irregularity for the discriminant class d15 indicatesthat this technique is isolated from other techniques

413 Calculating Complexity Structure complexity for asample with α3 24 is

Q 1 1 2( 1113857

ψ(K) 2(2 + 2 + 3)

(3lowast 4)1113890 1113891 117

(22)

e system complexity for different values of alpha isshown in Figure 12 As can be seen the total complexity ofthe system for large values alpha tends to stability

414 Ranking the Cryptography Algorithm ImplementationTechniques ese techniques can be classified into 5 levelsusing the analysis of the obtained results ese levels can beseen in Table 16 and Figure 13 e link power of the factorsin one group is specified with alpha Each level describespriority and importance of the group in the development ofcryptography algorithms Accordingly six methods of basicscience key management using hardware methods usingsteganography Avalanche effect and using hybrid method

q Q21o

112

SETSX15X15X1X11X15X19X27X13 amp X22

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-10-1-1-10-1-1-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-10-1-1-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

2-10-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1

-1-1-1-1-1-1

-1-1-1-1-1

-1-1-1-1

-1-1-1

-1-10

Figure 11 Results of executing the model for cryptography algorithm implementation techniques with α3 24

14 Computational Intelligence and Neuroscience

are the most important cryptography algorithm imple-mentation techniques to achieve the main objectives

5 Conclusion

In this study the design and implementation model of thecryptography algorithms is designed in three levels and itseffective components are extracted To organize the com-ponents in addition to elimination of the additional com-ponents through the measure reduction algorithm a properclassification is applied to examine the mutual effects Afterclassification three 13lowast 4 7lowast 4 and 29lowast 7 matrices areconstituted and the model is implemented on these threematrices To implement the designed model Q-analysis isused Accordingly for support indices five indices of highpriority include human resources research and develop-ment management organizationstructure and equipmentFor the algorithm design objectives index five high priorityindices include security speed optimal usage of resources

and simplicity For the indices related to implementationtechniques of the cryptography algorithms the most appliedtechniques for achieving the determined objectives includeusing basic science hardware and software methods usingkeymanagement hybrid method steganography Avalancheeffect and using artificial intelligence In the future work wewill consider the following cases

A plan should be formulated according to the prioritiesAccording to the outputs of this study and the presented pri-orities the following topics can be investigated in future studies

(1) Presenting a comprehensive model for generatingvarious cryptography algorithms based on the pri-orities of interest

(2) Examining and formulating a model about the roleof basic science for design and implementation ofcryptography algorithms considering the significantrole of ldquobasic sciencerdquo in implementation of thecryptography algorithms and about the role ofschools and universities

(3) Presenting a model for design and implementationof cryptography algorithms in the IoT infrastructurewith optimal resource usage

Data Availability

e data used to support the findings of the study areavailable from the corresponding author upon request

Conflicts of Interest

e authors declare that there are no conflicts of interestregarding the publication of this article

Table 15 Irregularity of the cryptography algorithm implementation techniques for α3 24

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d15 qi 2 1 3 2 1- qi 1 1 2 033

11117

1 1 1

091

1112

com

plex

ity

alpha-cutα=2 α=3 α=5 α=15 α=25

Figure 12 System complexity of cryptography algorithm implementation techniques for different alpha-cuts

Table 16 Ranking the implementation techniques of the cryptography algorithms considering various alpha-cuts using Q-analysis

Equivalence classes for the parameters with minimum relationship level (q 0) αBasic sciences α20 163Key α10 81Hardware and steganography α5 40Avalanche effect and hybrid method α3 24Artificial intelligence software occupation area energy consumption memory consumption α2 16

basic sciences

key

Avalanche effect hybrid method

hardware steganography

artificial intelligence soware occupationarea energy consumption memory consumption

Figure 13 Prioritizing the implementation techniques of thecryptography algorithms using Q-analysis

Computational Intelligence and Neuroscience 15

Acknowledgments

is work was partially supported by Iran National ScienceFoundation (INSF) under contract No 9653979

References

[1] D Nilesh and M Nagle ldquoe new cryptography algorithmwith high throughputrdquo in Proceedings of the 2014 Interna-tional Conference on Computer Communication and Infor-matics pp 3ndash7 Coimbatore India January 2014

[2] S Ravi P Kocher R Lee G McGraw and A RaghunathanldquoSecurity as a new dimension in embedded system designrdquo inProceedings of the 41st annual conference on Design auto-mation-DAC lsquo04 pp 753ndash760 San Diego CA USA March2004

[3] P Kuppuswamy and S Q Y A Khalidi ldquoHybrid encryptiondecryption technique using new public key and symmetric keyalgorithmrdquo International Journal of Information and Com-puter Security vol 6 no 4 pp 372ndash382 2014

[4] S Bhat and V Kapoor ldquoSecure and efficient data privacyauthentication and integrity schemes using hybrid cryptog-raphyrdquo International Conference on Advanced ComputingNetworking and Informatics vol 870 pp 279ndash285 2019

[5] D Ott and C Peikert ldquoAnd other workshop participantsldquoidentifying research challenges in post quantum cryptog-raphy migration and cryptographic agilityrdquo 2019 httparxivorgabs190907353

[6] P Manickam K Shankar E Perumal M Ilayaraja andK Sathesh Kumar ldquoSecure data transmission through reliablevehicles in VANET using optimal lightweight cryptographyrdquoin Advanced Sciences and Technologies for Security Applica-tions pp 193ndash204 Springer New York NY USA 2019

[7] S Prakash and A Rajput ldquoHybrid cryptography for securedata communication in wireless sensor networksrdquo Advancesin Intelligent Systems and Computing vol 696 pp 589ndash5992018

[8] A A Soofi I Riaz and U Rasheed ldquoAn enhanced vigenerecipher for data securityrdquo International Journal of Scientific ampTechnology Research vol 4 no 8 pp 141ndash145 2015

[9] N Shamsuddin M Syafiqah and S Ali Pitchay ldquoImple-menting location-based cryptography on mobile applicationdesign to secure data in cloud storagerdquo Journal of PhysicsConference Series IOP Publishing vol 1551 no 1 2020

[10] L Jiao H Yonglin and D Feng ldquoStream cipher designs areviewrdquo Science China Information Sciences vol 63 no 3pp 1ndash25 2020

[11] NIST ldquoNIST cryptographic standards and guidelines devel-opment processrdquo Inside NIST vol 27 2016

[12] G Alagic J Alperin-Sheriff D Apon et al Status Report onthe First Round of the NIST Post-Quantum CryptographyStandardization Process National Institute of Standards andTechnology Gaithersburg MD USA pp 1ndash27 2019

[13] I Damaj and S Kasbah ldquoAn analysis framework for hardwareand software implementations with applications from cryp-tographyrdquo Computers amp Electrical Engineering vol 69pp 572ndash584 2018

[14] M S Croock Z A Hassan and S D Khudhur ldquoAdaptive keygeneration algorithm based on software engineering meth-odologyrdquo International Journal of Electrical and ComputerEngineering vol 11 no 1 p 589 2021

[15] K Sha T A Yang W Wei and S Davari ldquoA survey of edgecomputing-based designs for iot securityrdquo Digital Commu-nications and Networks vol 6 no 2 pp 195ndash202 2020

[16] M Yahuza M Y I B Idris A W B A Wahab et alldquoSystematic review on security and privacy requirements inedge computing state of the art and future research oppor-tunitiesrdquo IEEE Access vol 8 pp 76541ndash76567 2020

[17] A Bhowmik J Dey and S Karforma A Novel Double TierCryptographic System (nDTCS) to Reinforce Patientsrsquo Privacyin Contemporary COVID-19 Telemedicine Springer NewYork NY USA 2021

[18] N Gupta User Integrity Protection Security Model forEnhanced Telemedicine for Healthcare Networks 2020httpsosfio69jdq

[19] J Dey A Bhowmik A Sarkar S Karforma andB Chowdhury Cryptographic Engineering on COVID-19Telemedicine An Intelligent Transmission through RecurrentRelation Based Session Key Springer New York NY USA2021

[20] V Hosseinian H Ayatollahi H Haghani and E MehraeenldquoRequirements of information security in a telemedicinenetwork review of IT managersrsquo opinionrdquo Journal of Para-medical Sciences amp Rehabilitation vol 4 no 2 pp 31ndash402015

[21] D Karaoglan Altop A Levi and V Tuzcu ldquoDeriving cryp-tographic keys from physiological signalsrdquo Pervasive andMobile Computing vol 39 pp 65ndash79 2017

[22] O Uzunkol and M S Kiraz ldquoStill wrong use of pairings incryptographyrdquo Applied Mathematics and Computationvol 333 pp 467ndash479 2018

[23] S Picek D Jakobovic J F Miller L Batina and M CupicldquoCryptographic Boolean functions one output many designcriteriardquo Applied Soft Computing vol 40 pp 635ndash653 2016

[24] R Saha and G Geetha ldquoSymmetric random function gen-erator (SRFG) a novel cryptographic primitive for designingfast and robust algorithmsrdquo Chaos Solitons amp Fractalsvol 104 pp 371ndash377 2017

[25] A Achuthshankar and A Achuthshankar ldquoA novel sym-metric cryptography algorithm for fast and secure encryp-tionrdquo in Proceedings of the 2015 IEEE 9th InternationalConference on Intelligent Systems and Control (ISCO) pp 4ndash9Coimbatore India January 2015

[26] D Ott and C Peikert ldquoIdentifying research challenges in postquantum cryptography migration and cryptographic agilityrdquo2019 httparxivorgabs190907353

[27] W Liu B Ying H Yang and H Wang ldquoAccurate modelingfor predicting cryptography overheads on wireless sensornodesrdquovol 2 pp 997ndash1001 in Proceedings of the 2019 11thInternational Conference on Advanced CommunicationTechnology vol 2 pp 997ndash1001 IEEE Bangalore IndiaFebruary 2019

[28] S Feizi A Ahmadi and A Nemati ldquoA hardware imple-mentation of simon cryptography algorithmrdquo in Proceedingsof the 2014 4th International Conference on Computer andKnowledge Engineering (ICCKE) pp 245ndash250 IEEEMashhad Iran 2014

[29] E ambiraja G Ramesh and R Umarani ldquoA survey onvarious most common encryption techniquesrdquo 2012 httpwwwijarcssecom

[30] A Gupta and N K Walia ldquoCryptography Algorithms areviewrdquo International Journal of Engineering Research andDevelopment vol 2 no 2 pp 1667ndash1672 2014

[31] R Bhanot and R Hans ldquoA review and comparative analysis ofvarious encryption algorithmsrdquo International Journal of Se-curity and Its Applications vol 9 no 4 pp 289ndash306 2015

[32] S Chandra S Paira S S Alam and G Sanyal ldquoA com-parative survey of symmetric and asymmetric key

16 Computational Intelligence and Neuroscience

cryptographyrdquo in Proceedings of the 2014 InternationalConference on Electronics Communication and Computa-tional Engineering (ICECCE) pp 83ndash93 Hosur India No-vember 2014

[33] F Maqsood M Ahmed M Mumtaz and M Ali ldquoCryp-tography a comparative analysis for modern techniquesrdquoInternational Journal of Advanced Computer Science andApplications vol 8 no 6 p 6 2017

[34] S B Sadkhan and A O Salman ldquoA survey on lightweight-cryptography status and future challengesrdquo in Proceedings ofthe International Conference on Advances in SustainableEngineering and Applications pp 105ndash108 Wasit Iraq June2018

[35] N V oai ldquoCriteria and dimension reduction of linearmultiple criteria optimization problemsrdquo Journal of GlobalOptimization vol 52 no 3 pp 499ndash508 2012

[36] C E Shannon ldquoCommunication theory of secrecy systems1945rdquo MD Computing vol 15 no 1 pp 57ndash64 1998

[37] L J Fennelly M Beaudry and M A Perry Security in 2025[38] ITU Global Cybersecurity Index 2018 2019[39] ITU Global Cybersecurity Index (GCI) 2017 2017

Computational Intelligence and Neuroscience 17

σ0 d1( 1113857

σminus1 d2( 1113857

σminus1 d3( 1113857

σminus1 d4( 1113857

σminus1 d5( 1113857

σminus1 d6( 1113857

σminus1 d7( 1113857

σminus1 d8( 1113857

σminus1 d9( 1113857

σminus1 d10( 1113857

σ0 d11( 1113857

σminus1 d12( 1113857

σ0 d13( 1113857

σminus1 d14( 1113857

σ2 d15( 1113857

σminus1 d16( 1113857

σminus1 d17( 1113857

σminus1 d18( 1113857

σ0 d19( 1113857

σminus1 d20( 1113857

σminus1 d21( 1113857

σ0 d22( 1113857

σminus1 d23( 1113857

σminus1 d24( 1113857

σminus1 d25( 1113857

σminus1 d26( 1113857

σ0 d27( 1113857

σminus1 d28( 1113857

σminus1 d29( 1113857

(17)

erefore the complex dimension is 2 In other wordsthe discriminant class d15 (basic sciences) has the largestdimension

49 Calculating the Dimensions and the Q-Link In the alphadefined for cryptography algorithm implementation tech-niques to achieve the defined goals the obtained q-linkshows that these techniques are relatively independent al-though there is a weak relationship between some tech-niques Q-link in the samples with α3 24 between eachtwo subsequent di is as follows

σ0 d1( 1113857 σminus1 d2( 1113857⟶ minus 1

σminus1 d2( 1113857 σminus1 d3( 1113857⟶ minus 1

σ0 d3( 1113857 σminus1 d4( 1113857⟶ minus 1

σminus1 d4( 1113857 σminus1 d5( 1113857⟶ minus 1

σminus1 d5( 1113857 σminus1 d6( 1113857⟶ minus 1

σminus1 d6( 1113857 σminus1 d7( 1113857⟶ minus 1

σminus1 d7( 1113857 σminus1 d8( 1113857⟶ minus 1

σminus1 d8( 1113857 σminus1 d9( 1113857⟶ minus 1

σminus1 d9( 1113857 σminus1 d10( 1113857⟶ minus 1

σminus1 d10( 1113857 σ0 d11( 1113857⟶ minus 1

σ0 d11( 1113857 σminus1 d12( 1113857⟶ minus 1

σminus1 d12( 1113857 σ0 d13( 1113857⟶ minus 1

σ0 d13( 1113857 σminus1 d14( 1113857⟶ minus 1

σminus1 d14( 1113857 σ2 d15( 1113857⟶ minus 1

σ2 d15( 1113857 σminus1 d16( 1113857⟶ minus 1

σminus1 d16( 1113857 σminus1 d17( 1113857⟶ minus 1

σminus1 d17( 1113857 σminus1 d18( 1113857⟶ minus 1

σminus1 d18( 1113857 σ0 d19( 1113857⟶ minus 1

σ0 d19( 1113857 σminus1 d20( 1113857⟶ minus 1

σminus1 d20( 1113857 σminus1 d21( 1113857⟶ minus 1

σminus1 d21( 1113857 σ0 d22( 1113857⟶ minus 1

σ0 d22( 1113857 σminus1 d23( 1113857⟶ minus 1

σminus1 d23( 1113857 σminus1 d24( 1113857⟶ minus 1

σminus1 d24( 1113857 σminus1 d25( 1113857⟶ minus 1

σminus1 d25( 1113857 σminus1 d26( 1113857⟶ minus 1

σminus1 d26( 1113857 σ0 d27( 1113857⟶ minus 1

σ0 d27( 1113857 σminus1 d28( 1113857⟶ minus 1

σminus1 d28( 1113857 σminus1 d29( 1113857⟶ minus 1

(18)

Table 11 Irregularity of the cryptography objectivesrsquo parameters in the data matrix A for α5

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d1 and d6 qi 3 2 and 1 433 3 072d3 qi 2 and 1 133 3 026d4 qi 1 033 3 006

11117

1 1 1

09095

1105

11115

12

com

plex

ity

alpha cutα=2 α=3 α=5 α=15 α=25

Figure 8 System complexity of the cryptography algorithm designobjectives for different alpha-cuts

12 Computational Intelligence and Neuroscience

410 Calculating the Structure Vectors erefore the firststructure vector based on the software output is

Q 1 1 2( 1113857 (19)

e second structure vector P is

P 1 1 29( 1113857 (20)

Since the large values of P of higher dimensions dem-onstrate more links the second structure vector calculatedfor the alpha of interest shows that the relationship betweenthe cryptography algorithm implementation techniques isminimum

411 Calculating the Obstruction or Inflexibility Vectore obstruction vector (Qlowast) for a sample with α3 24 is

Table 12 Ranking the objectivesrsquo parameters considering different cuts in Q-analysis

e equivalence class for the parameters with minimum relation level (q 0) αSecurity α50 244Speed α30 244Resources α15 122Flexibility α5 40Simplicity-scalability α2 16(All parameters)

securityspeed

optimal usage of resources

flexibility

simplicity-scalability

Figure 9 Prioritizing the design and implementation objectives of the cryptography algorithms using Q-analysis

securityspeed

optimal usage of resources

flexibility

simplicity-scalability-reliability

Figure 10 Prioritizing the design and implementation objectives of the cryptography algorithms using Q-analysis

Table 13 Set of dis and cis the techniques employed for implementing cryptography algorithms

Flexibility C4 Fuzzy logic d25 Storage space d17 Cycle d9 Avalanche effect d1Scalability C5 Software d26 Clustering d18 Multiple step d10 Digital signature d2Speed C6 Steganography d27 Key d19 Hybrid method d11 Block size d3Reliability C7 Music harmony d28 Graph d20 Hardware d12 Image sharing d4

Artificial intelligence d29 Characteristic oriented d21 Hardware d13 Parallel processing d5Security C1 Energy consumption d22 Occupation area d14 reshold technique d6Simplicity C2 Bandwidth consumption d23 Basic science d15 Data mining d7Resources C3 Memory consumption d24 Compression d16 Binary tree d8

Table 14 e incidence matrix of the cryptography algorithmimplementation techniques with α 24

C1 C2 C3 C4 C5 C6 C7

d1 1 0 0 0 0 0 0d2 0 0 0 0 0 0 0d3 0 0 0 0 0 0 0d4 0 0 0 0 0 0 0d5 0 0 0 0 0 0 0d6 0 0 0 0 0 0 0d7 0 0 0 0 0 0 0d8 0 0 0 0 0 0 0d9 0 0 0 0 0 0 0d10 0 0 0 0 0 0 0d11 1 0 0 0 0 0 0d12 0 0 0 0 0 0 0

Computational Intelligence and Neuroscience 13

Qlowast

Q minus I⟶ Qlowast

1 1 21113858 1113859 minus 1 1 11113858 1113859⟶ Qlowast

0 0 11113858 1113859

(21)

e value of QlowastK represents the number of structurallimitations or obstructions for the cryptography techniquesrsquointeraction at dimension k As can be seen from the cal-culated obstruction vector there is a significant relationshipat some levels

412 Calculating Irregularity Irregularity is the integrationdegree of a cryptography method in the total complex Mea-suring irregularity (eccrsquo) for α3 24 is shown in Table 15 Ascan be seen irregularity for the discriminant class d15 indicatesthat this technique is isolated from other techniques

413 Calculating Complexity Structure complexity for asample with α3 24 is

Q 1 1 2( 1113857

ψ(K) 2(2 + 2 + 3)

(3lowast 4)1113890 1113891 117

(22)

e system complexity for different values of alpha isshown in Figure 12 As can be seen the total complexity ofthe system for large values alpha tends to stability

414 Ranking the Cryptography Algorithm ImplementationTechniques ese techniques can be classified into 5 levelsusing the analysis of the obtained results ese levels can beseen in Table 16 and Figure 13 e link power of the factorsin one group is specified with alpha Each level describespriority and importance of the group in the development ofcryptography algorithms Accordingly six methods of basicscience key management using hardware methods usingsteganography Avalanche effect and using hybrid method

q Q21o

112

SETSX15X15X1X11X15X19X27X13 amp X22

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-10-1-1-10-1-1-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-10-1-1-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

2-10-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1

-1-1-1-1-1-1

-1-1-1-1-1

-1-1-1-1

-1-1-1

-1-10

Figure 11 Results of executing the model for cryptography algorithm implementation techniques with α3 24

14 Computational Intelligence and Neuroscience

are the most important cryptography algorithm imple-mentation techniques to achieve the main objectives

5 Conclusion

In this study the design and implementation model of thecryptography algorithms is designed in three levels and itseffective components are extracted To organize the com-ponents in addition to elimination of the additional com-ponents through the measure reduction algorithm a properclassification is applied to examine the mutual effects Afterclassification three 13lowast 4 7lowast 4 and 29lowast 7 matrices areconstituted and the model is implemented on these threematrices To implement the designed model Q-analysis isused Accordingly for support indices five indices of highpriority include human resources research and develop-ment management organizationstructure and equipmentFor the algorithm design objectives index five high priorityindices include security speed optimal usage of resources

and simplicity For the indices related to implementationtechniques of the cryptography algorithms the most appliedtechniques for achieving the determined objectives includeusing basic science hardware and software methods usingkeymanagement hybrid method steganography Avalancheeffect and using artificial intelligence In the future work wewill consider the following cases

A plan should be formulated according to the prioritiesAccording to the outputs of this study and the presented pri-orities the following topics can be investigated in future studies

(1) Presenting a comprehensive model for generatingvarious cryptography algorithms based on the pri-orities of interest

(2) Examining and formulating a model about the roleof basic science for design and implementation ofcryptography algorithms considering the significantrole of ldquobasic sciencerdquo in implementation of thecryptography algorithms and about the role ofschools and universities

(3) Presenting a model for design and implementationof cryptography algorithms in the IoT infrastructurewith optimal resource usage

Data Availability

e data used to support the findings of the study areavailable from the corresponding author upon request

Conflicts of Interest

e authors declare that there are no conflicts of interestregarding the publication of this article

Table 15 Irregularity of the cryptography algorithm implementation techniques for α3 24

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d15 qi 2 1 3 2 1- qi 1 1 2 033

11117

1 1 1

091

1112

com

plex

ity

alpha-cutα=2 α=3 α=5 α=15 α=25

Figure 12 System complexity of cryptography algorithm implementation techniques for different alpha-cuts

Table 16 Ranking the implementation techniques of the cryptography algorithms considering various alpha-cuts using Q-analysis

Equivalence classes for the parameters with minimum relationship level (q 0) αBasic sciences α20 163Key α10 81Hardware and steganography α5 40Avalanche effect and hybrid method α3 24Artificial intelligence software occupation area energy consumption memory consumption α2 16

basic sciences

key

Avalanche effect hybrid method

hardware steganography

artificial intelligence soware occupationarea energy consumption memory consumption

Figure 13 Prioritizing the implementation techniques of thecryptography algorithms using Q-analysis

Computational Intelligence and Neuroscience 15

Acknowledgments

is work was partially supported by Iran National ScienceFoundation (INSF) under contract No 9653979

References

[1] D Nilesh and M Nagle ldquoe new cryptography algorithmwith high throughputrdquo in Proceedings of the 2014 Interna-tional Conference on Computer Communication and Infor-matics pp 3ndash7 Coimbatore India January 2014

[2] S Ravi P Kocher R Lee G McGraw and A RaghunathanldquoSecurity as a new dimension in embedded system designrdquo inProceedings of the 41st annual conference on Design auto-mation-DAC lsquo04 pp 753ndash760 San Diego CA USA March2004

[3] P Kuppuswamy and S Q Y A Khalidi ldquoHybrid encryptiondecryption technique using new public key and symmetric keyalgorithmrdquo International Journal of Information and Com-puter Security vol 6 no 4 pp 372ndash382 2014

[4] S Bhat and V Kapoor ldquoSecure and efficient data privacyauthentication and integrity schemes using hybrid cryptog-raphyrdquo International Conference on Advanced ComputingNetworking and Informatics vol 870 pp 279ndash285 2019

[5] D Ott and C Peikert ldquoAnd other workshop participantsldquoidentifying research challenges in post quantum cryptog-raphy migration and cryptographic agilityrdquo 2019 httparxivorgabs190907353

[6] P Manickam K Shankar E Perumal M Ilayaraja andK Sathesh Kumar ldquoSecure data transmission through reliablevehicles in VANET using optimal lightweight cryptographyrdquoin Advanced Sciences and Technologies for Security Applica-tions pp 193ndash204 Springer New York NY USA 2019

[7] S Prakash and A Rajput ldquoHybrid cryptography for securedata communication in wireless sensor networksrdquo Advancesin Intelligent Systems and Computing vol 696 pp 589ndash5992018

[8] A A Soofi I Riaz and U Rasheed ldquoAn enhanced vigenerecipher for data securityrdquo International Journal of Scientific ampTechnology Research vol 4 no 8 pp 141ndash145 2015

[9] N Shamsuddin M Syafiqah and S Ali Pitchay ldquoImple-menting location-based cryptography on mobile applicationdesign to secure data in cloud storagerdquo Journal of PhysicsConference Series IOP Publishing vol 1551 no 1 2020

[10] L Jiao H Yonglin and D Feng ldquoStream cipher designs areviewrdquo Science China Information Sciences vol 63 no 3pp 1ndash25 2020

[11] NIST ldquoNIST cryptographic standards and guidelines devel-opment processrdquo Inside NIST vol 27 2016

[12] G Alagic J Alperin-Sheriff D Apon et al Status Report onthe First Round of the NIST Post-Quantum CryptographyStandardization Process National Institute of Standards andTechnology Gaithersburg MD USA pp 1ndash27 2019

[13] I Damaj and S Kasbah ldquoAn analysis framework for hardwareand software implementations with applications from cryp-tographyrdquo Computers amp Electrical Engineering vol 69pp 572ndash584 2018

[14] M S Croock Z A Hassan and S D Khudhur ldquoAdaptive keygeneration algorithm based on software engineering meth-odologyrdquo International Journal of Electrical and ComputerEngineering vol 11 no 1 p 589 2021

[15] K Sha T A Yang W Wei and S Davari ldquoA survey of edgecomputing-based designs for iot securityrdquo Digital Commu-nications and Networks vol 6 no 2 pp 195ndash202 2020

[16] M Yahuza M Y I B Idris A W B A Wahab et alldquoSystematic review on security and privacy requirements inedge computing state of the art and future research oppor-tunitiesrdquo IEEE Access vol 8 pp 76541ndash76567 2020

[17] A Bhowmik J Dey and S Karforma A Novel Double TierCryptographic System (nDTCS) to Reinforce Patientsrsquo Privacyin Contemporary COVID-19 Telemedicine Springer NewYork NY USA 2021

[18] N Gupta User Integrity Protection Security Model forEnhanced Telemedicine for Healthcare Networks 2020httpsosfio69jdq

[19] J Dey A Bhowmik A Sarkar S Karforma andB Chowdhury Cryptographic Engineering on COVID-19Telemedicine An Intelligent Transmission through RecurrentRelation Based Session Key Springer New York NY USA2021

[20] V Hosseinian H Ayatollahi H Haghani and E MehraeenldquoRequirements of information security in a telemedicinenetwork review of IT managersrsquo opinionrdquo Journal of Para-medical Sciences amp Rehabilitation vol 4 no 2 pp 31ndash402015

[21] D Karaoglan Altop A Levi and V Tuzcu ldquoDeriving cryp-tographic keys from physiological signalsrdquo Pervasive andMobile Computing vol 39 pp 65ndash79 2017

[22] O Uzunkol and M S Kiraz ldquoStill wrong use of pairings incryptographyrdquo Applied Mathematics and Computationvol 333 pp 467ndash479 2018

[23] S Picek D Jakobovic J F Miller L Batina and M CupicldquoCryptographic Boolean functions one output many designcriteriardquo Applied Soft Computing vol 40 pp 635ndash653 2016

[24] R Saha and G Geetha ldquoSymmetric random function gen-erator (SRFG) a novel cryptographic primitive for designingfast and robust algorithmsrdquo Chaos Solitons amp Fractalsvol 104 pp 371ndash377 2017

[25] A Achuthshankar and A Achuthshankar ldquoA novel sym-metric cryptography algorithm for fast and secure encryp-tionrdquo in Proceedings of the 2015 IEEE 9th InternationalConference on Intelligent Systems and Control (ISCO) pp 4ndash9Coimbatore India January 2015

[26] D Ott and C Peikert ldquoIdentifying research challenges in postquantum cryptography migration and cryptographic agilityrdquo2019 httparxivorgabs190907353

[27] W Liu B Ying H Yang and H Wang ldquoAccurate modelingfor predicting cryptography overheads on wireless sensornodesrdquovol 2 pp 997ndash1001 in Proceedings of the 2019 11thInternational Conference on Advanced CommunicationTechnology vol 2 pp 997ndash1001 IEEE Bangalore IndiaFebruary 2019

[28] S Feizi A Ahmadi and A Nemati ldquoA hardware imple-mentation of simon cryptography algorithmrdquo in Proceedingsof the 2014 4th International Conference on Computer andKnowledge Engineering (ICCKE) pp 245ndash250 IEEEMashhad Iran 2014

[29] E ambiraja G Ramesh and R Umarani ldquoA survey onvarious most common encryption techniquesrdquo 2012 httpwwwijarcssecom

[30] A Gupta and N K Walia ldquoCryptography Algorithms areviewrdquo International Journal of Engineering Research andDevelopment vol 2 no 2 pp 1667ndash1672 2014

[31] R Bhanot and R Hans ldquoA review and comparative analysis ofvarious encryption algorithmsrdquo International Journal of Se-curity and Its Applications vol 9 no 4 pp 289ndash306 2015

[32] S Chandra S Paira S S Alam and G Sanyal ldquoA com-parative survey of symmetric and asymmetric key

16 Computational Intelligence and Neuroscience

cryptographyrdquo in Proceedings of the 2014 InternationalConference on Electronics Communication and Computa-tional Engineering (ICECCE) pp 83ndash93 Hosur India No-vember 2014

[33] F Maqsood M Ahmed M Mumtaz and M Ali ldquoCryp-tography a comparative analysis for modern techniquesrdquoInternational Journal of Advanced Computer Science andApplications vol 8 no 6 p 6 2017

[34] S B Sadkhan and A O Salman ldquoA survey on lightweight-cryptography status and future challengesrdquo in Proceedings ofthe International Conference on Advances in SustainableEngineering and Applications pp 105ndash108 Wasit Iraq June2018

[35] N V oai ldquoCriteria and dimension reduction of linearmultiple criteria optimization problemsrdquo Journal of GlobalOptimization vol 52 no 3 pp 499ndash508 2012

[36] C E Shannon ldquoCommunication theory of secrecy systems1945rdquo MD Computing vol 15 no 1 pp 57ndash64 1998

[37] L J Fennelly M Beaudry and M A Perry Security in 2025[38] ITU Global Cybersecurity Index 2018 2019[39] ITU Global Cybersecurity Index (GCI) 2017 2017

Computational Intelligence and Neuroscience 17

410 Calculating the Structure Vectors erefore the firststructure vector based on the software output is

Q 1 1 2( 1113857 (19)

e second structure vector P is

P 1 1 29( 1113857 (20)

Since the large values of P of higher dimensions dem-onstrate more links the second structure vector calculatedfor the alpha of interest shows that the relationship betweenthe cryptography algorithm implementation techniques isminimum

411 Calculating the Obstruction or Inflexibility Vectore obstruction vector (Qlowast) for a sample with α3 24 is

Table 12 Ranking the objectivesrsquo parameters considering different cuts in Q-analysis

e equivalence class for the parameters with minimum relation level (q 0) αSecurity α50 244Speed α30 244Resources α15 122Flexibility α5 40Simplicity-scalability α2 16(All parameters)

securityspeed

optimal usage of resources

flexibility

simplicity-scalability

Figure 9 Prioritizing the design and implementation objectives of the cryptography algorithms using Q-analysis

securityspeed

optimal usage of resources

flexibility

simplicity-scalability-reliability

Figure 10 Prioritizing the design and implementation objectives of the cryptography algorithms using Q-analysis

Table 13 Set of dis and cis the techniques employed for implementing cryptography algorithms

Flexibility C4 Fuzzy logic d25 Storage space d17 Cycle d9 Avalanche effect d1Scalability C5 Software d26 Clustering d18 Multiple step d10 Digital signature d2Speed C6 Steganography d27 Key d19 Hybrid method d11 Block size d3Reliability C7 Music harmony d28 Graph d20 Hardware d12 Image sharing d4

Artificial intelligence d29 Characteristic oriented d21 Hardware d13 Parallel processing d5Security C1 Energy consumption d22 Occupation area d14 reshold technique d6Simplicity C2 Bandwidth consumption d23 Basic science d15 Data mining d7Resources C3 Memory consumption d24 Compression d16 Binary tree d8

Table 14 e incidence matrix of the cryptography algorithmimplementation techniques with α 24

C1 C2 C3 C4 C5 C6 C7

d1 1 0 0 0 0 0 0d2 0 0 0 0 0 0 0d3 0 0 0 0 0 0 0d4 0 0 0 0 0 0 0d5 0 0 0 0 0 0 0d6 0 0 0 0 0 0 0d7 0 0 0 0 0 0 0d8 0 0 0 0 0 0 0d9 0 0 0 0 0 0 0d10 0 0 0 0 0 0 0d11 1 0 0 0 0 0 0d12 0 0 0 0 0 0 0

Computational Intelligence and Neuroscience 13

Qlowast

Q minus I⟶ Qlowast

1 1 21113858 1113859 minus 1 1 11113858 1113859⟶ Qlowast

0 0 11113858 1113859

(21)

e value of QlowastK represents the number of structurallimitations or obstructions for the cryptography techniquesrsquointeraction at dimension k As can be seen from the cal-culated obstruction vector there is a significant relationshipat some levels

412 Calculating Irregularity Irregularity is the integrationdegree of a cryptography method in the total complex Mea-suring irregularity (eccrsquo) for α3 24 is shown in Table 15 Ascan be seen irregularity for the discriminant class d15 indicatesthat this technique is isolated from other techniques

413 Calculating Complexity Structure complexity for asample with α3 24 is

Q 1 1 2( 1113857

ψ(K) 2(2 + 2 + 3)

(3lowast 4)1113890 1113891 117

(22)

e system complexity for different values of alpha isshown in Figure 12 As can be seen the total complexity ofthe system for large values alpha tends to stability

414 Ranking the Cryptography Algorithm ImplementationTechniques ese techniques can be classified into 5 levelsusing the analysis of the obtained results ese levels can beseen in Table 16 and Figure 13 e link power of the factorsin one group is specified with alpha Each level describespriority and importance of the group in the development ofcryptography algorithms Accordingly six methods of basicscience key management using hardware methods usingsteganography Avalanche effect and using hybrid method

q Q21o

112

SETSX15X15X1X11X15X19X27X13 amp X22

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-10-1-1-10-1-1-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-10-1-1-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

2-10-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1

-1-1-1-1-1-1

-1-1-1-1-1

-1-1-1-1

-1-1-1

-1-10

Figure 11 Results of executing the model for cryptography algorithm implementation techniques with α3 24

14 Computational Intelligence and Neuroscience

are the most important cryptography algorithm imple-mentation techniques to achieve the main objectives

5 Conclusion

In this study the design and implementation model of thecryptography algorithms is designed in three levels and itseffective components are extracted To organize the com-ponents in addition to elimination of the additional com-ponents through the measure reduction algorithm a properclassification is applied to examine the mutual effects Afterclassification three 13lowast 4 7lowast 4 and 29lowast 7 matrices areconstituted and the model is implemented on these threematrices To implement the designed model Q-analysis isused Accordingly for support indices five indices of highpriority include human resources research and develop-ment management organizationstructure and equipmentFor the algorithm design objectives index five high priorityindices include security speed optimal usage of resources

and simplicity For the indices related to implementationtechniques of the cryptography algorithms the most appliedtechniques for achieving the determined objectives includeusing basic science hardware and software methods usingkeymanagement hybrid method steganography Avalancheeffect and using artificial intelligence In the future work wewill consider the following cases

A plan should be formulated according to the prioritiesAccording to the outputs of this study and the presented pri-orities the following topics can be investigated in future studies

(1) Presenting a comprehensive model for generatingvarious cryptography algorithms based on the pri-orities of interest

(2) Examining and formulating a model about the roleof basic science for design and implementation ofcryptography algorithms considering the significantrole of ldquobasic sciencerdquo in implementation of thecryptography algorithms and about the role ofschools and universities

(3) Presenting a model for design and implementationof cryptography algorithms in the IoT infrastructurewith optimal resource usage

Data Availability

e data used to support the findings of the study areavailable from the corresponding author upon request

Conflicts of Interest

e authors declare that there are no conflicts of interestregarding the publication of this article

Table 15 Irregularity of the cryptography algorithm implementation techniques for α3 24

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d15 qi 2 1 3 2 1- qi 1 1 2 033

11117

1 1 1

091

1112

com

plex

ity

alpha-cutα=2 α=3 α=5 α=15 α=25

Figure 12 System complexity of cryptography algorithm implementation techniques for different alpha-cuts

Table 16 Ranking the implementation techniques of the cryptography algorithms considering various alpha-cuts using Q-analysis

Equivalence classes for the parameters with minimum relationship level (q 0) αBasic sciences α20 163Key α10 81Hardware and steganography α5 40Avalanche effect and hybrid method α3 24Artificial intelligence software occupation area energy consumption memory consumption α2 16

basic sciences

key

Avalanche effect hybrid method

hardware steganography

artificial intelligence soware occupationarea energy consumption memory consumption

Figure 13 Prioritizing the implementation techniques of thecryptography algorithms using Q-analysis

Computational Intelligence and Neuroscience 15

Acknowledgments

is work was partially supported by Iran National ScienceFoundation (INSF) under contract No 9653979

References

[1] D Nilesh and M Nagle ldquoe new cryptography algorithmwith high throughputrdquo in Proceedings of the 2014 Interna-tional Conference on Computer Communication and Infor-matics pp 3ndash7 Coimbatore India January 2014

[2] S Ravi P Kocher R Lee G McGraw and A RaghunathanldquoSecurity as a new dimension in embedded system designrdquo inProceedings of the 41st annual conference on Design auto-mation-DAC lsquo04 pp 753ndash760 San Diego CA USA March2004

[3] P Kuppuswamy and S Q Y A Khalidi ldquoHybrid encryptiondecryption technique using new public key and symmetric keyalgorithmrdquo International Journal of Information and Com-puter Security vol 6 no 4 pp 372ndash382 2014

[4] S Bhat and V Kapoor ldquoSecure and efficient data privacyauthentication and integrity schemes using hybrid cryptog-raphyrdquo International Conference on Advanced ComputingNetworking and Informatics vol 870 pp 279ndash285 2019

[5] D Ott and C Peikert ldquoAnd other workshop participantsldquoidentifying research challenges in post quantum cryptog-raphy migration and cryptographic agilityrdquo 2019 httparxivorgabs190907353

[6] P Manickam K Shankar E Perumal M Ilayaraja andK Sathesh Kumar ldquoSecure data transmission through reliablevehicles in VANET using optimal lightweight cryptographyrdquoin Advanced Sciences and Technologies for Security Applica-tions pp 193ndash204 Springer New York NY USA 2019

[7] S Prakash and A Rajput ldquoHybrid cryptography for securedata communication in wireless sensor networksrdquo Advancesin Intelligent Systems and Computing vol 696 pp 589ndash5992018

[8] A A Soofi I Riaz and U Rasheed ldquoAn enhanced vigenerecipher for data securityrdquo International Journal of Scientific ampTechnology Research vol 4 no 8 pp 141ndash145 2015

[9] N Shamsuddin M Syafiqah and S Ali Pitchay ldquoImple-menting location-based cryptography on mobile applicationdesign to secure data in cloud storagerdquo Journal of PhysicsConference Series IOP Publishing vol 1551 no 1 2020

[10] L Jiao H Yonglin and D Feng ldquoStream cipher designs areviewrdquo Science China Information Sciences vol 63 no 3pp 1ndash25 2020

[11] NIST ldquoNIST cryptographic standards and guidelines devel-opment processrdquo Inside NIST vol 27 2016

[12] G Alagic J Alperin-Sheriff D Apon et al Status Report onthe First Round of the NIST Post-Quantum CryptographyStandardization Process National Institute of Standards andTechnology Gaithersburg MD USA pp 1ndash27 2019

[13] I Damaj and S Kasbah ldquoAn analysis framework for hardwareand software implementations with applications from cryp-tographyrdquo Computers amp Electrical Engineering vol 69pp 572ndash584 2018

[14] M S Croock Z A Hassan and S D Khudhur ldquoAdaptive keygeneration algorithm based on software engineering meth-odologyrdquo International Journal of Electrical and ComputerEngineering vol 11 no 1 p 589 2021

[15] K Sha T A Yang W Wei and S Davari ldquoA survey of edgecomputing-based designs for iot securityrdquo Digital Commu-nications and Networks vol 6 no 2 pp 195ndash202 2020

[16] M Yahuza M Y I B Idris A W B A Wahab et alldquoSystematic review on security and privacy requirements inedge computing state of the art and future research oppor-tunitiesrdquo IEEE Access vol 8 pp 76541ndash76567 2020

[17] A Bhowmik J Dey and S Karforma A Novel Double TierCryptographic System (nDTCS) to Reinforce Patientsrsquo Privacyin Contemporary COVID-19 Telemedicine Springer NewYork NY USA 2021

[18] N Gupta User Integrity Protection Security Model forEnhanced Telemedicine for Healthcare Networks 2020httpsosfio69jdq

[19] J Dey A Bhowmik A Sarkar S Karforma andB Chowdhury Cryptographic Engineering on COVID-19Telemedicine An Intelligent Transmission through RecurrentRelation Based Session Key Springer New York NY USA2021

[20] V Hosseinian H Ayatollahi H Haghani and E MehraeenldquoRequirements of information security in a telemedicinenetwork review of IT managersrsquo opinionrdquo Journal of Para-medical Sciences amp Rehabilitation vol 4 no 2 pp 31ndash402015

[21] D Karaoglan Altop A Levi and V Tuzcu ldquoDeriving cryp-tographic keys from physiological signalsrdquo Pervasive andMobile Computing vol 39 pp 65ndash79 2017

[22] O Uzunkol and M S Kiraz ldquoStill wrong use of pairings incryptographyrdquo Applied Mathematics and Computationvol 333 pp 467ndash479 2018

[23] S Picek D Jakobovic J F Miller L Batina and M CupicldquoCryptographic Boolean functions one output many designcriteriardquo Applied Soft Computing vol 40 pp 635ndash653 2016

[24] R Saha and G Geetha ldquoSymmetric random function gen-erator (SRFG) a novel cryptographic primitive for designingfast and robust algorithmsrdquo Chaos Solitons amp Fractalsvol 104 pp 371ndash377 2017

[25] A Achuthshankar and A Achuthshankar ldquoA novel sym-metric cryptography algorithm for fast and secure encryp-tionrdquo in Proceedings of the 2015 IEEE 9th InternationalConference on Intelligent Systems and Control (ISCO) pp 4ndash9Coimbatore India January 2015

[26] D Ott and C Peikert ldquoIdentifying research challenges in postquantum cryptography migration and cryptographic agilityrdquo2019 httparxivorgabs190907353

[27] W Liu B Ying H Yang and H Wang ldquoAccurate modelingfor predicting cryptography overheads on wireless sensornodesrdquovol 2 pp 997ndash1001 in Proceedings of the 2019 11thInternational Conference on Advanced CommunicationTechnology vol 2 pp 997ndash1001 IEEE Bangalore IndiaFebruary 2019

[28] S Feizi A Ahmadi and A Nemati ldquoA hardware imple-mentation of simon cryptography algorithmrdquo in Proceedingsof the 2014 4th International Conference on Computer andKnowledge Engineering (ICCKE) pp 245ndash250 IEEEMashhad Iran 2014

[29] E ambiraja G Ramesh and R Umarani ldquoA survey onvarious most common encryption techniquesrdquo 2012 httpwwwijarcssecom

[30] A Gupta and N K Walia ldquoCryptography Algorithms areviewrdquo International Journal of Engineering Research andDevelopment vol 2 no 2 pp 1667ndash1672 2014

[31] R Bhanot and R Hans ldquoA review and comparative analysis ofvarious encryption algorithmsrdquo International Journal of Se-curity and Its Applications vol 9 no 4 pp 289ndash306 2015

[32] S Chandra S Paira S S Alam and G Sanyal ldquoA com-parative survey of symmetric and asymmetric key

16 Computational Intelligence and Neuroscience

cryptographyrdquo in Proceedings of the 2014 InternationalConference on Electronics Communication and Computa-tional Engineering (ICECCE) pp 83ndash93 Hosur India No-vember 2014

[33] F Maqsood M Ahmed M Mumtaz and M Ali ldquoCryp-tography a comparative analysis for modern techniquesrdquoInternational Journal of Advanced Computer Science andApplications vol 8 no 6 p 6 2017

[34] S B Sadkhan and A O Salman ldquoA survey on lightweight-cryptography status and future challengesrdquo in Proceedings ofthe International Conference on Advances in SustainableEngineering and Applications pp 105ndash108 Wasit Iraq June2018

[35] N V oai ldquoCriteria and dimension reduction of linearmultiple criteria optimization problemsrdquo Journal of GlobalOptimization vol 52 no 3 pp 499ndash508 2012

[36] C E Shannon ldquoCommunication theory of secrecy systems1945rdquo MD Computing vol 15 no 1 pp 57ndash64 1998

[37] L J Fennelly M Beaudry and M A Perry Security in 2025[38] ITU Global Cybersecurity Index 2018 2019[39] ITU Global Cybersecurity Index (GCI) 2017 2017

Computational Intelligence and Neuroscience 17

Qlowast

Q minus I⟶ Qlowast

1 1 21113858 1113859 minus 1 1 11113858 1113859⟶ Qlowast

0 0 11113858 1113859

(21)

e value of QlowastK represents the number of structurallimitations or obstructions for the cryptography techniquesrsquointeraction at dimension k As can be seen from the cal-culated obstruction vector there is a significant relationshipat some levels

412 Calculating Irregularity Irregularity is the integrationdegree of a cryptography method in the total complex Mea-suring irregularity (eccrsquo) for α3 24 is shown in Table 15 Ascan be seen irregularity for the discriminant class d15 indicatesthat this technique is isolated from other techniques

413 Calculating Complexity Structure complexity for asample with α3 24 is

Q 1 1 2( 1113857

ψ(K) 2(2 + 2 + 3)

(3lowast 4)1113890 1113891 117

(22)

e system complexity for different values of alpha isshown in Figure 12 As can be seen the total complexity ofthe system for large values alpha tends to stability

414 Ranking the Cryptography Algorithm ImplementationTechniques ese techniques can be classified into 5 levelsusing the analysis of the obtained results ese levels can beseen in Table 16 and Figure 13 e link power of the factorsin one group is specified with alpha Each level describespriority and importance of the group in the development ofcryptography algorithms Accordingly six methods of basicscience key management using hardware methods usingsteganography Avalanche effect and using hybrid method

q Q21o

112

SETSX15X15X1X11X15X19X27X13 amp X22

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-10-1-1-10-1-1-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-10-1-1-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1

2-10-10-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1-1-1-1

0-1-1-1-1-1-1-1-1-10

-1-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1-1

-1-1-1-1-1-1-1

-1-1-1-1-1-1

-1-1-1-1-1

-1-1-1-1

-1-1-1

-1-10

Figure 11 Results of executing the model for cryptography algorithm implementation techniques with α3 24

14 Computational Intelligence and Neuroscience

are the most important cryptography algorithm imple-mentation techniques to achieve the main objectives

5 Conclusion

In this study the design and implementation model of thecryptography algorithms is designed in three levels and itseffective components are extracted To organize the com-ponents in addition to elimination of the additional com-ponents through the measure reduction algorithm a properclassification is applied to examine the mutual effects Afterclassification three 13lowast 4 7lowast 4 and 29lowast 7 matrices areconstituted and the model is implemented on these threematrices To implement the designed model Q-analysis isused Accordingly for support indices five indices of highpriority include human resources research and develop-ment management organizationstructure and equipmentFor the algorithm design objectives index five high priorityindices include security speed optimal usage of resources

and simplicity For the indices related to implementationtechniques of the cryptography algorithms the most appliedtechniques for achieving the determined objectives includeusing basic science hardware and software methods usingkeymanagement hybrid method steganography Avalancheeffect and using artificial intelligence In the future work wewill consider the following cases

A plan should be formulated according to the prioritiesAccording to the outputs of this study and the presented pri-orities the following topics can be investigated in future studies

(1) Presenting a comprehensive model for generatingvarious cryptography algorithms based on the pri-orities of interest

(2) Examining and formulating a model about the roleof basic science for design and implementation ofcryptography algorithms considering the significantrole of ldquobasic sciencerdquo in implementation of thecryptography algorithms and about the role ofschools and universities

(3) Presenting a model for design and implementationof cryptography algorithms in the IoT infrastructurewith optimal resource usage

Data Availability

e data used to support the findings of the study areavailable from the corresponding author upon request

Conflicts of Interest

e authors declare that there are no conflicts of interestregarding the publication of this article

Table 15 Irregularity of the cryptography algorithm implementation techniques for α3 24

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d15 qi 2 1 3 2 1- qi 1 1 2 033

11117

1 1 1

091

1112

com

plex

ity

alpha-cutα=2 α=3 α=5 α=15 α=25

Figure 12 System complexity of cryptography algorithm implementation techniques for different alpha-cuts

Table 16 Ranking the implementation techniques of the cryptography algorithms considering various alpha-cuts using Q-analysis

Equivalence classes for the parameters with minimum relationship level (q 0) αBasic sciences α20 163Key α10 81Hardware and steganography α5 40Avalanche effect and hybrid method α3 24Artificial intelligence software occupation area energy consumption memory consumption α2 16

basic sciences

key

Avalanche effect hybrid method

hardware steganography

artificial intelligence soware occupationarea energy consumption memory consumption

Figure 13 Prioritizing the implementation techniques of thecryptography algorithms using Q-analysis

Computational Intelligence and Neuroscience 15

Acknowledgments

is work was partially supported by Iran National ScienceFoundation (INSF) under contract No 9653979

References

[1] D Nilesh and M Nagle ldquoe new cryptography algorithmwith high throughputrdquo in Proceedings of the 2014 Interna-tional Conference on Computer Communication and Infor-matics pp 3ndash7 Coimbatore India January 2014

[2] S Ravi P Kocher R Lee G McGraw and A RaghunathanldquoSecurity as a new dimension in embedded system designrdquo inProceedings of the 41st annual conference on Design auto-mation-DAC lsquo04 pp 753ndash760 San Diego CA USA March2004

[3] P Kuppuswamy and S Q Y A Khalidi ldquoHybrid encryptiondecryption technique using new public key and symmetric keyalgorithmrdquo International Journal of Information and Com-puter Security vol 6 no 4 pp 372ndash382 2014

[4] S Bhat and V Kapoor ldquoSecure and efficient data privacyauthentication and integrity schemes using hybrid cryptog-raphyrdquo International Conference on Advanced ComputingNetworking and Informatics vol 870 pp 279ndash285 2019

[5] D Ott and C Peikert ldquoAnd other workshop participantsldquoidentifying research challenges in post quantum cryptog-raphy migration and cryptographic agilityrdquo 2019 httparxivorgabs190907353

[6] P Manickam K Shankar E Perumal M Ilayaraja andK Sathesh Kumar ldquoSecure data transmission through reliablevehicles in VANET using optimal lightweight cryptographyrdquoin Advanced Sciences and Technologies for Security Applica-tions pp 193ndash204 Springer New York NY USA 2019

[7] S Prakash and A Rajput ldquoHybrid cryptography for securedata communication in wireless sensor networksrdquo Advancesin Intelligent Systems and Computing vol 696 pp 589ndash5992018

[8] A A Soofi I Riaz and U Rasheed ldquoAn enhanced vigenerecipher for data securityrdquo International Journal of Scientific ampTechnology Research vol 4 no 8 pp 141ndash145 2015

[9] N Shamsuddin M Syafiqah and S Ali Pitchay ldquoImple-menting location-based cryptography on mobile applicationdesign to secure data in cloud storagerdquo Journal of PhysicsConference Series IOP Publishing vol 1551 no 1 2020

[10] L Jiao H Yonglin and D Feng ldquoStream cipher designs areviewrdquo Science China Information Sciences vol 63 no 3pp 1ndash25 2020

[11] NIST ldquoNIST cryptographic standards and guidelines devel-opment processrdquo Inside NIST vol 27 2016

[12] G Alagic J Alperin-Sheriff D Apon et al Status Report onthe First Round of the NIST Post-Quantum CryptographyStandardization Process National Institute of Standards andTechnology Gaithersburg MD USA pp 1ndash27 2019

[13] I Damaj and S Kasbah ldquoAn analysis framework for hardwareand software implementations with applications from cryp-tographyrdquo Computers amp Electrical Engineering vol 69pp 572ndash584 2018

[14] M S Croock Z A Hassan and S D Khudhur ldquoAdaptive keygeneration algorithm based on software engineering meth-odologyrdquo International Journal of Electrical and ComputerEngineering vol 11 no 1 p 589 2021

[15] K Sha T A Yang W Wei and S Davari ldquoA survey of edgecomputing-based designs for iot securityrdquo Digital Commu-nications and Networks vol 6 no 2 pp 195ndash202 2020

[16] M Yahuza M Y I B Idris A W B A Wahab et alldquoSystematic review on security and privacy requirements inedge computing state of the art and future research oppor-tunitiesrdquo IEEE Access vol 8 pp 76541ndash76567 2020

[17] A Bhowmik J Dey and S Karforma A Novel Double TierCryptographic System (nDTCS) to Reinforce Patientsrsquo Privacyin Contemporary COVID-19 Telemedicine Springer NewYork NY USA 2021

[18] N Gupta User Integrity Protection Security Model forEnhanced Telemedicine for Healthcare Networks 2020httpsosfio69jdq

[19] J Dey A Bhowmik A Sarkar S Karforma andB Chowdhury Cryptographic Engineering on COVID-19Telemedicine An Intelligent Transmission through RecurrentRelation Based Session Key Springer New York NY USA2021

[20] V Hosseinian H Ayatollahi H Haghani and E MehraeenldquoRequirements of information security in a telemedicinenetwork review of IT managersrsquo opinionrdquo Journal of Para-medical Sciences amp Rehabilitation vol 4 no 2 pp 31ndash402015

[21] D Karaoglan Altop A Levi and V Tuzcu ldquoDeriving cryp-tographic keys from physiological signalsrdquo Pervasive andMobile Computing vol 39 pp 65ndash79 2017

[22] O Uzunkol and M S Kiraz ldquoStill wrong use of pairings incryptographyrdquo Applied Mathematics and Computationvol 333 pp 467ndash479 2018

[23] S Picek D Jakobovic J F Miller L Batina and M CupicldquoCryptographic Boolean functions one output many designcriteriardquo Applied Soft Computing vol 40 pp 635ndash653 2016

[24] R Saha and G Geetha ldquoSymmetric random function gen-erator (SRFG) a novel cryptographic primitive for designingfast and robust algorithmsrdquo Chaos Solitons amp Fractalsvol 104 pp 371ndash377 2017

[25] A Achuthshankar and A Achuthshankar ldquoA novel sym-metric cryptography algorithm for fast and secure encryp-tionrdquo in Proceedings of the 2015 IEEE 9th InternationalConference on Intelligent Systems and Control (ISCO) pp 4ndash9Coimbatore India January 2015

[26] D Ott and C Peikert ldquoIdentifying research challenges in postquantum cryptography migration and cryptographic agilityrdquo2019 httparxivorgabs190907353

[27] W Liu B Ying H Yang and H Wang ldquoAccurate modelingfor predicting cryptography overheads on wireless sensornodesrdquovol 2 pp 997ndash1001 in Proceedings of the 2019 11thInternational Conference on Advanced CommunicationTechnology vol 2 pp 997ndash1001 IEEE Bangalore IndiaFebruary 2019

[28] S Feizi A Ahmadi and A Nemati ldquoA hardware imple-mentation of simon cryptography algorithmrdquo in Proceedingsof the 2014 4th International Conference on Computer andKnowledge Engineering (ICCKE) pp 245ndash250 IEEEMashhad Iran 2014

[29] E ambiraja G Ramesh and R Umarani ldquoA survey onvarious most common encryption techniquesrdquo 2012 httpwwwijarcssecom

[30] A Gupta and N K Walia ldquoCryptography Algorithms areviewrdquo International Journal of Engineering Research andDevelopment vol 2 no 2 pp 1667ndash1672 2014

[31] R Bhanot and R Hans ldquoA review and comparative analysis ofvarious encryption algorithmsrdquo International Journal of Se-curity and Its Applications vol 9 no 4 pp 289ndash306 2015

[32] S Chandra S Paira S S Alam and G Sanyal ldquoA com-parative survey of symmetric and asymmetric key

16 Computational Intelligence and Neuroscience

cryptographyrdquo in Proceedings of the 2014 InternationalConference on Electronics Communication and Computa-tional Engineering (ICECCE) pp 83ndash93 Hosur India No-vember 2014

[33] F Maqsood M Ahmed M Mumtaz and M Ali ldquoCryp-tography a comparative analysis for modern techniquesrdquoInternational Journal of Advanced Computer Science andApplications vol 8 no 6 p 6 2017

[34] S B Sadkhan and A O Salman ldquoA survey on lightweight-cryptography status and future challengesrdquo in Proceedings ofthe International Conference on Advances in SustainableEngineering and Applications pp 105ndash108 Wasit Iraq June2018

[35] N V oai ldquoCriteria and dimension reduction of linearmultiple criteria optimization problemsrdquo Journal of GlobalOptimization vol 52 no 3 pp 499ndash508 2012

[36] C E Shannon ldquoCommunication theory of secrecy systems1945rdquo MD Computing vol 15 no 1 pp 57ndash64 1998

[37] L J Fennelly M Beaudry and M A Perry Security in 2025[38] ITU Global Cybersecurity Index 2018 2019[39] ITU Global Cybersecurity Index (GCI) 2017 2017

Computational Intelligence and Neuroscience 17

are the most important cryptography algorithm imple-mentation techniques to achieve the main objectives

5 Conclusion

In this study the design and implementation model of thecryptography algorithms is designed in three levels and itseffective components are extracted To organize the com-ponents in addition to elimination of the additional com-ponents through the measure reduction algorithm a properclassification is applied to examine the mutual effects Afterclassification three 13lowast 4 7lowast 4 and 29lowast 7 matrices areconstituted and the model is implemented on these threematrices To implement the designed model Q-analysis isused Accordingly for support indices five indices of highpriority include human resources research and develop-ment management organizationstructure and equipmentFor the algorithm design objectives index five high priorityindices include security speed optimal usage of resources

and simplicity For the indices related to implementationtechniques of the cryptography algorithms the most appliedtechniques for achieving the determined objectives includeusing basic science hardware and software methods usingkeymanagement hybrid method steganography Avalancheeffect and using artificial intelligence In the future work wewill consider the following cases

A plan should be formulated according to the prioritiesAccording to the outputs of this study and the presented pri-orities the following topics can be investigated in future studies

(1) Presenting a comprehensive model for generatingvarious cryptography algorithms based on the pri-orities of interest

(2) Examining and formulating a model about the roleof basic science for design and implementation ofcryptography algorithms considering the significantrole of ldquobasic sciencerdquo in implementation of thecryptography algorithms and about the role ofschools and universities

(3) Presenting a model for design and implementationof cryptography algorithms in the IoT infrastructurewith optimal resource usage

Data Availability

e data used to support the findings of the study areavailable from the corresponding author upon request

Conflicts of Interest

e authors declare that there are no conflicts of interestregarding the publication of this article

Table 15 Irregularity of the cryptography algorithm implementation techniques for α3 24

σ qi 1113936 qiσi qmax eccprime(σ) 21113936 qiσiqmax(qmax + 1)

d15 qi 2 1 3 2 1- qi 1 1 2 033

11117

1 1 1

091

1112

com

plex

ity

alpha-cutα=2 α=3 α=5 α=15 α=25

Figure 12 System complexity of cryptography algorithm implementation techniques for different alpha-cuts

Table 16 Ranking the implementation techniques of the cryptography algorithms considering various alpha-cuts using Q-analysis

Equivalence classes for the parameters with minimum relationship level (q 0) αBasic sciences α20 163Key α10 81Hardware and steganography α5 40Avalanche effect and hybrid method α3 24Artificial intelligence software occupation area energy consumption memory consumption α2 16

basic sciences

key

Avalanche effect hybrid method

hardware steganography

artificial intelligence soware occupationarea energy consumption memory consumption

Figure 13 Prioritizing the implementation techniques of thecryptography algorithms using Q-analysis

Computational Intelligence and Neuroscience 15

Acknowledgments

is work was partially supported by Iran National ScienceFoundation (INSF) under contract No 9653979

References

[1] D Nilesh and M Nagle ldquoe new cryptography algorithmwith high throughputrdquo in Proceedings of the 2014 Interna-tional Conference on Computer Communication and Infor-matics pp 3ndash7 Coimbatore India January 2014

[2] S Ravi P Kocher R Lee G McGraw and A RaghunathanldquoSecurity as a new dimension in embedded system designrdquo inProceedings of the 41st annual conference on Design auto-mation-DAC lsquo04 pp 753ndash760 San Diego CA USA March2004

[3] P Kuppuswamy and S Q Y A Khalidi ldquoHybrid encryptiondecryption technique using new public key and symmetric keyalgorithmrdquo International Journal of Information and Com-puter Security vol 6 no 4 pp 372ndash382 2014

[4] S Bhat and V Kapoor ldquoSecure and efficient data privacyauthentication and integrity schemes using hybrid cryptog-raphyrdquo International Conference on Advanced ComputingNetworking and Informatics vol 870 pp 279ndash285 2019

[5] D Ott and C Peikert ldquoAnd other workshop participantsldquoidentifying research challenges in post quantum cryptog-raphy migration and cryptographic agilityrdquo 2019 httparxivorgabs190907353

[6] P Manickam K Shankar E Perumal M Ilayaraja andK Sathesh Kumar ldquoSecure data transmission through reliablevehicles in VANET using optimal lightweight cryptographyrdquoin Advanced Sciences and Technologies for Security Applica-tions pp 193ndash204 Springer New York NY USA 2019

[7] S Prakash and A Rajput ldquoHybrid cryptography for securedata communication in wireless sensor networksrdquo Advancesin Intelligent Systems and Computing vol 696 pp 589ndash5992018

[8] A A Soofi I Riaz and U Rasheed ldquoAn enhanced vigenerecipher for data securityrdquo International Journal of Scientific ampTechnology Research vol 4 no 8 pp 141ndash145 2015

[9] N Shamsuddin M Syafiqah and S Ali Pitchay ldquoImple-menting location-based cryptography on mobile applicationdesign to secure data in cloud storagerdquo Journal of PhysicsConference Series IOP Publishing vol 1551 no 1 2020

[10] L Jiao H Yonglin and D Feng ldquoStream cipher designs areviewrdquo Science China Information Sciences vol 63 no 3pp 1ndash25 2020

[11] NIST ldquoNIST cryptographic standards and guidelines devel-opment processrdquo Inside NIST vol 27 2016

[12] G Alagic J Alperin-Sheriff D Apon et al Status Report onthe First Round of the NIST Post-Quantum CryptographyStandardization Process National Institute of Standards andTechnology Gaithersburg MD USA pp 1ndash27 2019

[13] I Damaj and S Kasbah ldquoAn analysis framework for hardwareand software implementations with applications from cryp-tographyrdquo Computers amp Electrical Engineering vol 69pp 572ndash584 2018

[14] M S Croock Z A Hassan and S D Khudhur ldquoAdaptive keygeneration algorithm based on software engineering meth-odologyrdquo International Journal of Electrical and ComputerEngineering vol 11 no 1 p 589 2021

[15] K Sha T A Yang W Wei and S Davari ldquoA survey of edgecomputing-based designs for iot securityrdquo Digital Commu-nications and Networks vol 6 no 2 pp 195ndash202 2020

[16] M Yahuza M Y I B Idris A W B A Wahab et alldquoSystematic review on security and privacy requirements inedge computing state of the art and future research oppor-tunitiesrdquo IEEE Access vol 8 pp 76541ndash76567 2020

[17] A Bhowmik J Dey and S Karforma A Novel Double TierCryptographic System (nDTCS) to Reinforce Patientsrsquo Privacyin Contemporary COVID-19 Telemedicine Springer NewYork NY USA 2021

[18] N Gupta User Integrity Protection Security Model forEnhanced Telemedicine for Healthcare Networks 2020httpsosfio69jdq

[19] J Dey A Bhowmik A Sarkar S Karforma andB Chowdhury Cryptographic Engineering on COVID-19Telemedicine An Intelligent Transmission through RecurrentRelation Based Session Key Springer New York NY USA2021

[20] V Hosseinian H Ayatollahi H Haghani and E MehraeenldquoRequirements of information security in a telemedicinenetwork review of IT managersrsquo opinionrdquo Journal of Para-medical Sciences amp Rehabilitation vol 4 no 2 pp 31ndash402015

[21] D Karaoglan Altop A Levi and V Tuzcu ldquoDeriving cryp-tographic keys from physiological signalsrdquo Pervasive andMobile Computing vol 39 pp 65ndash79 2017

[22] O Uzunkol and M S Kiraz ldquoStill wrong use of pairings incryptographyrdquo Applied Mathematics and Computationvol 333 pp 467ndash479 2018

[23] S Picek D Jakobovic J F Miller L Batina and M CupicldquoCryptographic Boolean functions one output many designcriteriardquo Applied Soft Computing vol 40 pp 635ndash653 2016

[24] R Saha and G Geetha ldquoSymmetric random function gen-erator (SRFG) a novel cryptographic primitive for designingfast and robust algorithmsrdquo Chaos Solitons amp Fractalsvol 104 pp 371ndash377 2017

[25] A Achuthshankar and A Achuthshankar ldquoA novel sym-metric cryptography algorithm for fast and secure encryp-tionrdquo in Proceedings of the 2015 IEEE 9th InternationalConference on Intelligent Systems and Control (ISCO) pp 4ndash9Coimbatore India January 2015

[26] D Ott and C Peikert ldquoIdentifying research challenges in postquantum cryptography migration and cryptographic agilityrdquo2019 httparxivorgabs190907353

[27] W Liu B Ying H Yang and H Wang ldquoAccurate modelingfor predicting cryptography overheads on wireless sensornodesrdquovol 2 pp 997ndash1001 in Proceedings of the 2019 11thInternational Conference on Advanced CommunicationTechnology vol 2 pp 997ndash1001 IEEE Bangalore IndiaFebruary 2019

[28] S Feizi A Ahmadi and A Nemati ldquoA hardware imple-mentation of simon cryptography algorithmrdquo in Proceedingsof the 2014 4th International Conference on Computer andKnowledge Engineering (ICCKE) pp 245ndash250 IEEEMashhad Iran 2014

[29] E ambiraja G Ramesh and R Umarani ldquoA survey onvarious most common encryption techniquesrdquo 2012 httpwwwijarcssecom

[30] A Gupta and N K Walia ldquoCryptography Algorithms areviewrdquo International Journal of Engineering Research andDevelopment vol 2 no 2 pp 1667ndash1672 2014

[31] R Bhanot and R Hans ldquoA review and comparative analysis ofvarious encryption algorithmsrdquo International Journal of Se-curity and Its Applications vol 9 no 4 pp 289ndash306 2015

[32] S Chandra S Paira S S Alam and G Sanyal ldquoA com-parative survey of symmetric and asymmetric key

16 Computational Intelligence and Neuroscience

cryptographyrdquo in Proceedings of the 2014 InternationalConference on Electronics Communication and Computa-tional Engineering (ICECCE) pp 83ndash93 Hosur India No-vember 2014

[33] F Maqsood M Ahmed M Mumtaz and M Ali ldquoCryp-tography a comparative analysis for modern techniquesrdquoInternational Journal of Advanced Computer Science andApplications vol 8 no 6 p 6 2017

[34] S B Sadkhan and A O Salman ldquoA survey on lightweight-cryptography status and future challengesrdquo in Proceedings ofthe International Conference on Advances in SustainableEngineering and Applications pp 105ndash108 Wasit Iraq June2018

[35] N V oai ldquoCriteria and dimension reduction of linearmultiple criteria optimization problemsrdquo Journal of GlobalOptimization vol 52 no 3 pp 499ndash508 2012

[36] C E Shannon ldquoCommunication theory of secrecy systems1945rdquo MD Computing vol 15 no 1 pp 57ndash64 1998

[37] L J Fennelly M Beaudry and M A Perry Security in 2025[38] ITU Global Cybersecurity Index 2018 2019[39] ITU Global Cybersecurity Index (GCI) 2017 2017

Computational Intelligence and Neuroscience 17

Acknowledgments

is work was partially supported by Iran National ScienceFoundation (INSF) under contract No 9653979

References

[1] D Nilesh and M Nagle ldquoe new cryptography algorithmwith high throughputrdquo in Proceedings of the 2014 Interna-tional Conference on Computer Communication and Infor-matics pp 3ndash7 Coimbatore India January 2014

[2] S Ravi P Kocher R Lee G McGraw and A RaghunathanldquoSecurity as a new dimension in embedded system designrdquo inProceedings of the 41st annual conference on Design auto-mation-DAC lsquo04 pp 753ndash760 San Diego CA USA March2004

[3] P Kuppuswamy and S Q Y A Khalidi ldquoHybrid encryptiondecryption technique using new public key and symmetric keyalgorithmrdquo International Journal of Information and Com-puter Security vol 6 no 4 pp 372ndash382 2014

[4] S Bhat and V Kapoor ldquoSecure and efficient data privacyauthentication and integrity schemes using hybrid cryptog-raphyrdquo International Conference on Advanced ComputingNetworking and Informatics vol 870 pp 279ndash285 2019

[5] D Ott and C Peikert ldquoAnd other workshop participantsldquoidentifying research challenges in post quantum cryptog-raphy migration and cryptographic agilityrdquo 2019 httparxivorgabs190907353

[6] P Manickam K Shankar E Perumal M Ilayaraja andK Sathesh Kumar ldquoSecure data transmission through reliablevehicles in VANET using optimal lightweight cryptographyrdquoin Advanced Sciences and Technologies for Security Applica-tions pp 193ndash204 Springer New York NY USA 2019

[7] S Prakash and A Rajput ldquoHybrid cryptography for securedata communication in wireless sensor networksrdquo Advancesin Intelligent Systems and Computing vol 696 pp 589ndash5992018

[8] A A Soofi I Riaz and U Rasheed ldquoAn enhanced vigenerecipher for data securityrdquo International Journal of Scientific ampTechnology Research vol 4 no 8 pp 141ndash145 2015

[9] N Shamsuddin M Syafiqah and S Ali Pitchay ldquoImple-menting location-based cryptography on mobile applicationdesign to secure data in cloud storagerdquo Journal of PhysicsConference Series IOP Publishing vol 1551 no 1 2020

[10] L Jiao H Yonglin and D Feng ldquoStream cipher designs areviewrdquo Science China Information Sciences vol 63 no 3pp 1ndash25 2020

[11] NIST ldquoNIST cryptographic standards and guidelines devel-opment processrdquo Inside NIST vol 27 2016

[12] G Alagic J Alperin-Sheriff D Apon et al Status Report onthe First Round of the NIST Post-Quantum CryptographyStandardization Process National Institute of Standards andTechnology Gaithersburg MD USA pp 1ndash27 2019

[13] I Damaj and S Kasbah ldquoAn analysis framework for hardwareand software implementations with applications from cryp-tographyrdquo Computers amp Electrical Engineering vol 69pp 572ndash584 2018

[14] M S Croock Z A Hassan and S D Khudhur ldquoAdaptive keygeneration algorithm based on software engineering meth-odologyrdquo International Journal of Electrical and ComputerEngineering vol 11 no 1 p 589 2021

[15] K Sha T A Yang W Wei and S Davari ldquoA survey of edgecomputing-based designs for iot securityrdquo Digital Commu-nications and Networks vol 6 no 2 pp 195ndash202 2020

[16] M Yahuza M Y I B Idris A W B A Wahab et alldquoSystematic review on security and privacy requirements inedge computing state of the art and future research oppor-tunitiesrdquo IEEE Access vol 8 pp 76541ndash76567 2020

[17] A Bhowmik J Dey and S Karforma A Novel Double TierCryptographic System (nDTCS) to Reinforce Patientsrsquo Privacyin Contemporary COVID-19 Telemedicine Springer NewYork NY USA 2021

[18] N Gupta User Integrity Protection Security Model forEnhanced Telemedicine for Healthcare Networks 2020httpsosfio69jdq

[19] J Dey A Bhowmik A Sarkar S Karforma andB Chowdhury Cryptographic Engineering on COVID-19Telemedicine An Intelligent Transmission through RecurrentRelation Based Session Key Springer New York NY USA2021

[20] V Hosseinian H Ayatollahi H Haghani and E MehraeenldquoRequirements of information security in a telemedicinenetwork review of IT managersrsquo opinionrdquo Journal of Para-medical Sciences amp Rehabilitation vol 4 no 2 pp 31ndash402015

[21] D Karaoglan Altop A Levi and V Tuzcu ldquoDeriving cryp-tographic keys from physiological signalsrdquo Pervasive andMobile Computing vol 39 pp 65ndash79 2017

[22] O Uzunkol and M S Kiraz ldquoStill wrong use of pairings incryptographyrdquo Applied Mathematics and Computationvol 333 pp 467ndash479 2018

[23] S Picek D Jakobovic J F Miller L Batina and M CupicldquoCryptographic Boolean functions one output many designcriteriardquo Applied Soft Computing vol 40 pp 635ndash653 2016

[24] R Saha and G Geetha ldquoSymmetric random function gen-erator (SRFG) a novel cryptographic primitive for designingfast and robust algorithmsrdquo Chaos Solitons amp Fractalsvol 104 pp 371ndash377 2017

[25] A Achuthshankar and A Achuthshankar ldquoA novel sym-metric cryptography algorithm for fast and secure encryp-tionrdquo in Proceedings of the 2015 IEEE 9th InternationalConference on Intelligent Systems and Control (ISCO) pp 4ndash9Coimbatore India January 2015

[26] D Ott and C Peikert ldquoIdentifying research challenges in postquantum cryptography migration and cryptographic agilityrdquo2019 httparxivorgabs190907353

[27] W Liu B Ying H Yang and H Wang ldquoAccurate modelingfor predicting cryptography overheads on wireless sensornodesrdquovol 2 pp 997ndash1001 in Proceedings of the 2019 11thInternational Conference on Advanced CommunicationTechnology vol 2 pp 997ndash1001 IEEE Bangalore IndiaFebruary 2019

[28] S Feizi A Ahmadi and A Nemati ldquoA hardware imple-mentation of simon cryptography algorithmrdquo in Proceedingsof the 2014 4th International Conference on Computer andKnowledge Engineering (ICCKE) pp 245ndash250 IEEEMashhad Iran 2014

[29] E ambiraja G Ramesh and R Umarani ldquoA survey onvarious most common encryption techniquesrdquo 2012 httpwwwijarcssecom

[30] A Gupta and N K Walia ldquoCryptography Algorithms areviewrdquo International Journal of Engineering Research andDevelopment vol 2 no 2 pp 1667ndash1672 2014

[31] R Bhanot and R Hans ldquoA review and comparative analysis ofvarious encryption algorithmsrdquo International Journal of Se-curity and Its Applications vol 9 no 4 pp 289ndash306 2015

[32] S Chandra S Paira S S Alam and G Sanyal ldquoA com-parative survey of symmetric and asymmetric key

16 Computational Intelligence and Neuroscience

cryptographyrdquo in Proceedings of the 2014 InternationalConference on Electronics Communication and Computa-tional Engineering (ICECCE) pp 83ndash93 Hosur India No-vember 2014

[33] F Maqsood M Ahmed M Mumtaz and M Ali ldquoCryp-tography a comparative analysis for modern techniquesrdquoInternational Journal of Advanced Computer Science andApplications vol 8 no 6 p 6 2017

[34] S B Sadkhan and A O Salman ldquoA survey on lightweight-cryptography status and future challengesrdquo in Proceedings ofthe International Conference on Advances in SustainableEngineering and Applications pp 105ndash108 Wasit Iraq June2018

[35] N V oai ldquoCriteria and dimension reduction of linearmultiple criteria optimization problemsrdquo Journal of GlobalOptimization vol 52 no 3 pp 499ndash508 2012

[36] C E Shannon ldquoCommunication theory of secrecy systems1945rdquo MD Computing vol 15 no 1 pp 57ndash64 1998

[37] L J Fennelly M Beaudry and M A Perry Security in 2025[38] ITU Global Cybersecurity Index 2018 2019[39] ITU Global Cybersecurity Index (GCI) 2017 2017

Computational Intelligence and Neuroscience 17

cryptographyrdquo in Proceedings of the 2014 InternationalConference on Electronics Communication and Computa-tional Engineering (ICECCE) pp 83ndash93 Hosur India No-vember 2014

[33] F Maqsood M Ahmed M Mumtaz and M Ali ldquoCryp-tography a comparative analysis for modern techniquesrdquoInternational Journal of Advanced Computer Science andApplications vol 8 no 6 p 6 2017

[34] S B Sadkhan and A O Salman ldquoA survey on lightweight-cryptography status and future challengesrdquo in Proceedings ofthe International Conference on Advances in SustainableEngineering and Applications pp 105ndash108 Wasit Iraq June2018

[35] N V oai ldquoCriteria and dimension reduction of linearmultiple criteria optimization problemsrdquo Journal of GlobalOptimization vol 52 no 3 pp 499ndash508 2012

[36] C E Shannon ldquoCommunication theory of secrecy systems1945rdquo MD Computing vol 15 no 1 pp 57ndash64 1998

[37] L J Fennelly M Beaudry and M A Perry Security in 2025[38] ITU Global Cybersecurity Index 2018 2019[39] ITU Global Cybersecurity Index (GCI) 2017 2017

Computational Intelligence and Neuroscience 17