Research ArticleConstruction of Wireless Underground Footwork MobileTraining and Monitoring Sensor Network in Venues of MajorSports Events

Yang Wen1,2 and Fangliang Yu 2,3

1College of Sports Industry and Leisure, Nanjing Sport Institute, Nanjing 210014, China2Center of Jiangsu Sports Health Engineering Collaborative Innovation, Nanjing Sport Institute, Nanjing 210014, China3School of Sports Training, Nanjing Sport Institute, Nanjing 210014, China

Correspondence should be addressed to Fangliang Yu; 9120180005@nsi.edu.cn

Received 6 August 2021; Revised 20 August 2021; Accepted 21 August 2021; Published 31 August 2021

Academic Editor: Guolong Shi

Copyright © 2021 Yang Wen and Fangliang Yu. This is an open access article distributed under the Creative CommonsAttribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original workis properly cited.

At the Summer Olympics in Tokyo, technology was used extensively in major sports events. The level of foot movement abilitygreatly affects the performance of sports technology. Modern sports are developing in the direction of high speed, high skills,flexibility, and rapidity, and more and more reflect the important position of reasonable and accurate foot movement ability insports. This article uses wireless sensor technology and wireless communication technology to design the overall architecture ofthe wireless underground footwork mobile training and monitoring network in venues of major sports events. According tothe determined monitoring parameters and data transmission plan, a wireless remote monitoring data acquisition system isdesigned, and the hardware design, software design, and networking of the wireless monitoring node are completed, so as torealize the real-time monitoring and remote transmission of the underlying data. This paper proposes a wireless sensornetwork management architecture and method based on multiagent cooperation and combines active and passive wirelessunderground footwork mobile training and monitoring for experimental verification. A multitask allocation strategy optimizedfor network working life is proposed. A genetic algorithm is used to model and optimize the task data report routing of clusterhead nodes. The simulation experiment results show that the wireless sensor network management method based onmultiagent cooperation can effectively coordinate different monitoring sensor nodes to complete the assigned monitoring tasks;the multitask assignment strategy based on a genetic algorithm can optimize the working life of the application network.

1. Introduction

Footwork is the carrier of techniques and tactics. When thelegwork is effectively implemented, avoiding the opponent’soffense to make corresponding defenses, the footworkbecomes the basis for linking each technical action [1]. Ath-letes can adjust their own closed positions through the foot-work or open style to ensure that they are on a certainadvantage, and they can also use footwork ingeniously tomake corresponding adjustments in response to suddenattacks. When the athletes are in a confrontation on the field,the leg technique is equivalent, and when the tactics areequivalent, the athletes can effectively connect the various

links of the game and the technique and tactics by movingquickly and sensitively, so as to implement their own techni-cal level [2–4]. Therefore, whether to have a high level offootwork mobility will directly determine the athlete’s com-petition results. There is an inseparable connection betweenthe movement of footwork and the performance of the game[5]. The strong mobility of footwork has a certain impact onthe performance of athletes’ skills and tactics and the perfor-mance of the game. Footwork is the carrier of skills and tac-tics and an important guarantee for victory in the game.Flexible footwork movement is the key factor to achieve“integration of technology and tactics” and “integration ofoffense, defense, and counterattack.” To break through the

HindawiJournal of SensorsVolume 2021, Article ID 8423297, 11 pageshttps://doi.org/10.1155/2021/8423297

performance of high-level athletes, the combination of foot-work and legwork must be effectively improved, and tacticsmust be used throughout the game through footwork [6].

With the increase of national power, the number oflarge-scale events held has increased, and the number oflarge-scale events hosted by the world has also increased[7]. The requirements for the promptness, efficiency, andstability of information transmission are also increasing.Large-scale stadiums are often part of the central area ofthe event, as well as part of the construction of mobile com-munications. The central area is composed of several venuesand supporting auxiliary buildings [8]. To solve the contin-uous coverage of the central area and the venues is the mostbasic requirement. The most important thing is the deepcoverage of the central area of the large venues. With thegeneral popularization of 5G communication technology,the traditional standards for measuring wireless networkquality have also undergone major changes from the previ-ous 3G and 4G networks [9]. It has become very importantto discover and solve the impact of users’ perception of 4Guse through wireless network optimization methods. Theso-called user perception optimization improves the wirelessnetwork coverage quality, such as coverage rate, signalstrength, upload and download rate, and other indicatorsto meet the personalized needs of users, such as instant com-munication services, video services, and game services.Through perception optimization, it can improve user satis-faction during use and, on the other hand, tap the potentialof existing resources to maximize network resource effi-ciency, improve resource utilization and efficiency, andenhance the competitive advantage between operators [10].

This paper studies the collection and transmission ofhealth monitoring data and designs a wireless remote accesshealth monitoring system. Using sensor technology andwireless communication technology, we complete the overallarchitecture design of the health monitoring system andthen determine the main monitoring parameters accordingto the content of the structural health monitoring and designand develop the hardware system and software system of thehealth system. Specifically, the technical contributions of thisarticle can be summarized as follows:

(1) The debugging of the wireless monitoring node hasbeen completed, and the real-time collection of mon-itoring data from the sensors at the bottom of themonitoring system has been realized

(2) This article proposes a wireless sensor network man-agement architecture and method based on multia-gent cooperation and uses active and passivestructural health monitoring as an example to con-duct experimental verification

(3) The distributed underground footwork training andmonitoring method of wireless sensor network isstudied, and a multitask allocation strategy opti-mized for network working life is proposed

(4) We use a genetic algorithm to model and optimizethe task data reporting route of cluster head nodes.

The experimental test results show that, comparedwith the direct transmission of the star network,the multitask optimization of the method in thispaper can greatly improve the working life of thenetwork

2. Related Work

Relevant scholars pointed out that footwork, as a linkbetween offense and defense in actual combat, is anunchanging law through the ages [11]. Footwork skills areconcentrated on foot movement. In specific competitions,flexible foot movements can be effectively adjusted. The dis-tance between the opponent further realizes the offensiveand defensive skills through footwork, which directly affectsthe success or failure of the game. In actual competitions,when encountering opponents that are difficult to deal with,when the skills and tactics are equivalent, the athlete ana-lyzes the characteristics of the opponent, observes the oppo-nent’s body movements, and responds through footwork[12]. At this time, the ability of footwork movement is veryimportant. The definition of athlete’s mobility is mainly todefine athlete’s mobility in actual situations from the aspectsof athlete’s movement speed and reaction ability. Relevantscholars pointed out that mobility is the ability of athletesto initiate and complete specific actions in the shortest timeduring actual competitions [13]. The composing factors ofthis project’s ability are mainly composed of two parts: reac-tion ability and movement speed ability. Researchers believethat mobility refers to the ability of the batter to obtain thebest hitting point through the movement of footwork [14].It also requires athletes to choose and apply flexibly andaccurately according to the on-the-spot situation in a con-stantly changing competition environment.

Due to the uneven distribution of base stations, toomany base stations in a local area will cause overcoverageand even coverage confusion [15]. However, there are fewerbase stations in some areas, which will result in areas withweak coverage. The so-called overcoverage refers to the phe-nomenon of multiple network coverages in a certain area,which is also called cross-area coverage. Confusion of cover-age is caused by network interference due to too many net-works. Weak coverage refers to the phenomenon of weaknetwork coverage in a certain area. In response to the aboveproblems, using coverage optimization technology, the sys-tem can perform radio frequency optimization and optimi-zation of related parameters according to the locationenvironment of the area and the characteristics of the basestation, thereby controlling the coverage of the base stationantenna and reducing the occurrence of the above events.In LTE networks, hard handover is used to complete thehandover of mobile users between base stations [16, 17].The handover involved includes intrafrequency handover,interfrequency handover, and different system reselection.When a user moves to a different location, the network willbe handed over according to the neighboring cell selection toensure good signal quality and network speed. Therefore,through handover optimization technology, network hand-over can be realized, thereby ensuring network quality.

2 Journal of Sensors

However, in the handover process, the number of networksshould be appropriately selected, and the handover thresh-old should be set to ensure that the quality of the wirelessnetwork after handover optimization can meet the optimiza-tion goal [18].

In the early stage of LTE construction, the layout of basestations was carried out according to the physical location ofthe area [19]. However, after it is built, there may be a prob-lem of insufficient capacity caused by too many users. Inaddition, the reason for the mobility of mobile terminalusers is that the user is in a mobile state and the location isnot fixed, which causes the problem of unbalanced capacityof surrounding base stations. Therefore, the use of load bal-ancing optimization technology can achieve a balancedadjustment of traffic. The principle is to solve the capacityproblem according to the cell resources and traffic capacity,through the setting of handover and reselection parametersand through the increase of FDD-LTE 2100M equipmentand TDD-LTE equipment, so as to achieve the load balan-cing technology for different carriers. The traffic capacity isbalanced to finally achieve the optimization goal of the wire-less network and improve the network quality and user per-ception [20–22]. Currently, test instruments and relatedsoftware can be used to discover the range of network inter-ference and then use interference optimization technology tomake statistics on the network layout in the area and analyzethe causes of interference to formulate measures to optimizeLTE wireless network optimization to ensure network qual-ity. At present, the principle of interference optimizationtechnology is to maximize the reduction of mutual interfer-ence between base stations while ensuring that all areas arecovered by the network, so as to achieve the goal of improv-ing the quality of the base station network [23–25].

3. Design of the Health Monitoring System forVenues of Major Sports Events

3.1. Overall Design. Most large-scale space structures arelarge in scale, complex in structural design, and remote. Itis not suitable to use on-site wiring methods to obtain datacollected by sensors. The main reason is that there are manymonitoring points arranged, and on-site cable wiring notonly is difficult but also has a relatively high cost. Second,the wiring is messy and complicated, which is not conduciveto later management and maintenance, and there are hiddensafety hazards. Therefore, this article combines wireless sen-sor network (WSN) technology and wireless communicationtechnology, takes venues of major sports events as theresearch object, uses advanced health monitoring methodsto design the overall health monitoring system, and estab-lishes a set suitable for the long-term spatial structure. Aneffective health monitoring system ensures the monitoringof the entire process of construction, operation, mainte-nance, and repair of the stadium. Adopting the managementsystem of “distributed monitoring, centralized and hierar-chical management,” the entire health monitoring systemis mainly composed of three parts: the perception layer, thecommunication layer, and the management layer. The over-

all health monitoring system design architecture of the sys-tem is shown in Figure 1.

In the design of the health monitoring system, the bot-tom layer of the monitoring system is the perception layer,which is composed of WSN. The main task is to collectreal-time health monitoring data in the monitoring area.It is the basic core part of the entire health monitoringsystem. With their own advantages, wireless sensor net-works are widely used in various engineering structuralhealth monitoring fields. They have the characteristics oflow power consumption, self-organization and multihop,and strong network transmission reliability, which effec-tively solves the problems caused by traditional cable lay-outs. The problem solves the problems of large-spanspatial structure data transmission, sensor energy supply,and later maintenance and management, laying the foun-dation for future large-scale spatial structure health moni-toring and providing a new set of health monitoringtechnical means. When designing the perception layer inthis paper, ZigBee technology is used to build the entirewireless sensor network, which has flexible networkingand low power consumption functions, and then deployscorresponding sensors in the key parts of the steel struc-ture canopy of large-scale sports venues, such as stresssensors, temperature sensors, displacement sensors, windspeed, and direction sensors; these sensors can communi-cate normally with wireless communication devices, so asto realize the analysis, storage, and preprocessing of mon-itoring data and finally complete the ZigBee network con-figuration. It can be seen that the perception layer is thecore and basic part of the entire health monitoring system.So, this article adopts the ZigBee structure to completefunctions such as data collection of wireless monitoringnodes.

The communication layer is the middle layer of themonitoring system. It is mainly responsible for the conver-sion of the data transmission protocol of the monitoring sys-tem. It is the key to realizing the wireless remote healthmonitoring system. It mainly includes the ZigBee communi-cation network, embedded gateway, and GPRS communica-tion network. In order to meet the long-distancetransmission of monitoring data, this article uses the embed-ded gateway to convert the ZigBee communication protocolto the GPRS communication protocol and realizes the con-version of the wireless monitoring data communication pro-tocol. It can be said that the GPRS communication networkis a bridge for data transmission between the perceptionlayer and the management layer. In the entire health moni-toring data transmission process, the data collected by thesensor passes through the embedded gateway of the commu-nication layer, and the data collected by the ZigBee node isconnected to the GPRS network through the GPRS modulein the form of IP and then uploaded to the Web of the Inter-net by the GPRS base station. In the server, the monitoringdata collection and transmission process is completed. Withthe help of switches and routers, the monitored data isuploaded to the health monitoring information managementsystem to complete the real-time display, storage, and man-agement of the monitoring data.

3Journal of Sensors

The management layer is the management center of themonitoring system. It can provide management personnelwith monitoring information of each measuring point inthe monitoring system and provide a platform for the nextstep of intelligent analysis of monitoring data, intelligentdiagnosis of the structure, and hierarchical warning promptsof the system. This article adopts the form of IoT-level archi-tecture to design the overall health monitoring system archi-tecture, and it can meet the decentralized monitoring oflarge-scale spatial structures in different monitoring areas.Since traditional detection technology cannot meet the cur-rent detection requirements, in order to achieve real-timemonitoring of data, it is necessary to further systematic, sci-entific, and intelligent processing of real-time monitoringdata and finally provide preevaluation diagnosis results andtransformations for the monitored objects.

3.2. Monitoring System Hardware Design. Since most largespace structures are in the natural environment all yearround, they are susceptible to the adverse effects of environ-mental factors, leading to the early degradation or failure ofthe performance of the microprocessor and sensor monitor-ing nodes used in the monitoring system, which affects theentire wireless remote health monitoring system. Therefore,in the field of practical engineering applications, the designand research of wireless monitoring nodes and the selectionof hardware are very important, which determine the stabil-ity and scalability of the entire monitoring system. Figure 2shows the hardware composition of the wireless under-ground footwork mobile training and monitoring systemin venues of major sports events. This mainly uses ZigBeetechnology and GPRS technology to complete the remotetransmission of wireless sensor monitoring node data and

finally displays the monitoring data collected by the sensorin the health monitoring system in real time.

3.3. Wireless Communication Program Design. When thewireless communication module transmits the data collectedby the sensor to the sink node via the ZigBee router module,the coordinator needs to be networked. This is because thecoordinator module is always in a working state after poweron. If there is a communication request, the module willimmediately send instructions to the router module to makethe router module continuously send requests until the coor-dinator module responds. When the coordinator is network-ing, it needs to initialize the hardware configuration,protocol stack, network configuration, and external inter-face. After the initialization is completed, the coordinatormodule will keep the network monitoring and waiting stateuntil the monitoring node has a network access requestand then allocate the network address of the monitoringnode and update the information of the neighbor node,and finally, the coordinator module will perform thereceived data packet, determine whether the received infor-mation is data information, and discard if it is not therequired data packet.

When the router module sends monitoring data to thecoordinator module, it needs to initialize the network designof the router module. After the initial setting is completed,the router module is associated with the coordinator mod-ule, and then, a network access beacon request is sent tothe network. After the coordinator responds, it needs todetermine the response of the connection. If the coordinatorresponds correctly, it means that the binding address is cor-rect and the coordinator is successfully connected to the net-work; otherwise, the network connection fails.

Monitoring area 1

Monitoring area n

Stress sensorAccelerometerSpeed sensor

Motion detector

Stress sensorAccelerometerSpeed sensor

Motion detector

Wireless monitoring node

Wireless monitoring node

Embedded Gateway

Embedded Gateway

Base station

Internet

Web serverDatabaseLarge-scale sports venue structure

health monitoring platformLarge-scale sports venue structure health

monitoring information management system

Perception layer

Communication layer

Management layer

Wirelessmonitoring sensor

Wirelesscommunication

node

Zig Beecommunication

networkEmbedded

gatewayGPRS

communicationnetwork

Health monitoringdatabase

Health monitoringinformation

management system

Figure 1: Overall design architecture diagram of the health monitoring system of venues of major sports events.

4 Journal of Sensors

When the router module is successfully connected to thenetwork, when sending data, it will call the data processingsubroutine request to wake up the low power consumptionmode of the router module and then package the data infor-mation collected by the sensor and send it byte by byte. If therouter does not find the data sending command, it willalways be in a low-power sleep state. Only when there is adata sending command inside or outside, the router returnsto the working state from the dormant state and begins tocomplete the task processing.

4. Multitask Allocation Strategy for FootworkMovement Training and Monitoring

4.1. Wireless Sensor Network Collaboration Method UsingMultiagent Collaboration Technology. The mobile subject isa program that moves to the node with the requiredresources for calculation according to its own goals and theconditions of the required resources and can interact withother agents or resources; it can greatly reduce the data traf-fic in the network. It runs autonomously, encapsulates thetasks to be completed in the mobile agent, and dispatchesthem through the network. After that, the connectionbetween the source node and the target node can even bedisconnected. Therefore, the mobile agent has strong resil-ience, and fault tolerance is conducive to parallel distributedprocessing. Due to the characteristics and advantages ofmobile agents, it is very suitable for wireless sensor networksused in large-scale underground footwork mobile trainingand monitoring applications.

In SHM and test monitoring applications, wireless sen-sor network nodes often monitor the structural strain, vibra-tion, displacement, etc., in different areas in the form of

clustered subnets. Through local signal processing andregional information fusion, the structural feature datarequired for different monitoring tasks are obtained. There-fore, the wireless sensor network management architectureproposed in this paper includes three main types of networknodes: sensor nodes, cluster head nodes, and base stationmanagement nodes.

The sensor node has the functions of sensor data collec-tion, local signal processing, and wireless communication. Itis the smallest unit that forms a clustered network and canbecome the cluster head of the subnet according to differentmonitoring application requirements. The cluster head nodeis selected from the sensor nodes to be responsible for datafusion and forwarding tasks. In time-sensitive applications,it is also responsible for managing the synchronization oper-ation of local subnet monitoring. The base station manage-ment node is connected to the user monitoring terminalequipment; has unlimited processing, storage, and powersupply performance; and is responsible for the division, dis-tribution, and management of different application tasks.

This paper combines the application requirements ofwireless sensor networks for underground footwork mobiletraining and monitoring and designs a wireless sensor net-work management architecture based on multiagent cooper-ation, as shown in Figure 3. The framework extension of thisarticle defines the following six types of software main body:underground footwork mobile training monitoring mainbody, data management main body, monitoring applicationmain body, interpretation main body, central coordinationmain body, and user interface main body.

The main body of underground footwork mobile train-ing and monitoring refers to the main body of software usedto obtain structural data, which resides in each sensor node;

Debug interface

Clock circuit

Reset circuit

WirelesscommunicationZigBee module

WirelesscommunicationGPRS module

RS232 interface

Power supply

Communicationmodule RF interface

Embedded gateway hardwareblock diagram

Stress sensor

Accelerometer

Speed sensor

Motion detector

Sensor module

A/ Dconversion

CC2530processor GPRS module Network

indicator

Processor moduleWireless communication

module

Monitoring node hardware block diagram

Energy supply

Figure 2: The hardware composition of the wireless underground footwork mobile training and monitoring system in venues of majorsports events.

5Journal of Sensors

in active structural health monitoring, the main body alsoprovides driving signals for active structural excitation. Thedata management body is used to manage and mine the sen-sor data acquired by the underground footwork mobiletraining and monitoring body and is also responsible forthe clustering of sensor nodes and the routing of networkdata. The monitoring application subject is the softwaremobile subject directly connected with the central coordinat-ing subject and is responsible for the distribution, integra-tion, and migration of specific monitoring tasks. Itssoftware implementation includes four components: iden-tity, data space, fusion method, and migration path, whichare used for the mobile identification of the subject, storageof intermediate results, selection of fusion method, anddetermination of migration path. The main body of inter-pretation is the main body of software function that trans-forms specific application task operations into operationsperformed by local nodes. The existence of this softwaremain body makes the network system have good scalabilityand heterogeneous compatibility characteristics. The centralcoordinator distributes monitoring tasks through mobileagents and integrates the intermediate results of differentSAAs to get the most reliable and accurate conclusions.The main body of the user interface is responsible forobtaining monitoring instructions from users, forwardingand displaying monitoring conclusion data, etc.

4.2. Multitask Assignment Method Based on GeneticAlgorithm. In the multitask allocation process of under-ground footwork mobile training and monitoring, the mainenergy consumption process of the cluster head is in thereport sending stage of the task data. Therefore, the multi-task allocation of underground footwork mobile training

and monitoring can be simplified to the optimization ofthe task data report route for the cluster head node. If thereare n cluster head nodes in the network, and on average eachnode has d one-hop neighbor nodes (all cluster head nodes),then there are dn possibilities for possible task data reportingroutes. When the network scale is large, the number of suchrouting schemes will be massive. Therefore, this paperchooses the genetic algorithm which belongs to the heuristicsearch technology to find the optimal multitask assignmentscheme.

We establish a list Ni from n to 1, 1 ≤ i ≤ n, representingthe set of all neighboring nodes j of cluster head node i; here,the link i⟶ j ∈Ni is used to indicate that cluster head nodei passes through node j to the destination node routing. Inthis way, the initial population constitutes multiple genesand initial chromosomes by randomly selecting j in Ni.

After the initial population is generated, the fitness func-tion value of each individual needs to be calculated; becausethe goal of the underground footwork movement trainingmonitoring multitask allocation described in this article isto maximize the working life of the network, according tothe definition of the working life of the network, the life fit-ness function is defined as

Lnet =Eint

Emax − Emin − Eint: ð1Þ

Lnet is the network working life expressed in rounds, Einiis the initial energy of each cluster head node, and Emax is theenergy consumption value of the cluster head node with thelargest energy consumption during a round of task alloca-tion and execution. Here, the calculation of the sending

Data managemententity

Monitoringapplication subject

Monitoringsubject

Interpreter

Data managemententity

Monitoringapplication subject

Monitoringsubject

Interpreter

Sensor nodes (SNs)Sensor nodes (SNs)

Sensor nodes (SNs)

Sensor nodes (SNs)

Interpreter

Data management entity

Monitoring application subject

Monitoring the migration of application subjects

Wireless underground footwork mobile monitoring area in large scale sports venues-

Data managemententity

Monitoringsubject

Interpreter

Data managemententity

Monitoringsubject

Interpreter

Monitoring application subject Central coordinator User interface body Base Station Management Node (SMN)

Cluster headnodes (CHs)

Figure 3: Wireless sensor network management architecture based on multiagent cooperation.

6 Journal of Sensors

and receiving energy consumption of the cluster head nodeadopts the first-order wireless transceiver model:

ETi dij, bti� �

= αTbtidij,m − 1 − βð Þbti,ERi brið Þ = 1 − aRð Þ ⋅ bri − 1ð Þ:

ð2Þ

Among them, ETi represents the energy consumed bythe ith node to send data of size bti in a round of task alloca-tion; dij represents the Euclidean distance between nodes iand j; αT is the transmission energy coefficient, β is theamplifier coefficient, and the path loss index m is generally3-6; ERi represents the energy consumed by the ith node toreceive data of size bri in a round of task allocation; αR isthe received energy coefficient. Therefore, the node energyconsumption considered in this paper is the sum of theenergy consumption of sending and receiving.

In this paper, the rotation method is adopted to realizethe selection of individuals. According to the different valuesof the fitness function of the working life of the individuals,the probability of the individuals being selected is also differ-ent, which is a proportional selection strategy. We calculatethe working life fitness value LnetðiÞ for each chromosome

of the initial population and calculate the product of the fit-ness values of all chromosomes in the population:

ψ =Y

Lnet ið Þ: ð3Þ

For each chromosome, we calculate the probability ofselection:

p ið Þ = Lnet ið Þψ

: ð4Þ

Table 1: The main body and its functions in the collaborative management method.

Subject name Function description

Structural monitoring agent(SMA)

Responsible for inferring the location of the concentrated load applied to the structure by monitoring thechanges in the static load strain distribution

Data management agent(DMA)

Responsible for the moving average and zero compensation of the sensor monitoring data

Monitoring applicationagent (SAA)

Based on Euclid’s pattern recognition method to identify the strain pattern when the load is loaded todifferent positions

Interpretation agent (TA) Realize the mapping of software main functions to hardware execution operations based on TinyOS

Central coordinating agent(CCA)

Responsible for fusing the intermediate results of the pattern recognition method to obtain the judgmentresult of the loading position

User interface agent (UIA)Responsible for displaying monitoring network topology, monitoring intermediate results and final

judgment results, etc.

10 20 30 40 50 60 70 80 90 1000

2

4

6

8

10

12

14

16

Piezo sensor node

Tim

e-co

nsum

ing

mon

itorin

g/m

s

Figure 4: Time-consuming monitoring of wireless underground footwork mobile training.

Table 2: Simulation test parameter settings.

Test parameters Set value

The amount of data per task 1000 bits

Amplifier coefficient of wireless transceiver 85 pJ/bit/m2

The path loss index of the wireless transceiver 3~6Transmit power coefficient of wireless transceiver 35 nJ/bit

Received power coefficient of wireless transceiver 55 nJ/bit

Initial energy 3 J

7Journal of Sensors

For each chromosome, we calculate the cumulativeprobability:

q ið Þ =Y

p ið Þ: ð5Þ

In order to produce new offspring from the selected pop-ulation, this paper adopts the multipoint crossover methodto realize the reorganization of individuals. In the multipointcrossover method, there are m crossover points, and thecrossover position Ki (1 ≤ i ≤ n − 1, n is the length of thechromosome) is randomly generated in ascending order,and then, the genes between the parent chromosomes Kiare crossed to produce 2 new offspring chromosomes.

5. Simulation Experiment and Analysis

5.1. Simulation Experiment of Wireless Sensor NetworkCollaborative Management Method. Strain distributionmonitoring is to monitor the subject’s strain output throughunderground footwork movement training to monitor thestatic load position. In the experiment, the strain distribu-tion in the slab structure was changed by applying a concen-trated load, and the applied concentrated load was 55N. Ineach subarea, the output of the four-way underground foot-work mobile training monitoring body reflects the strain dis-tribution of the subsystem. When the position of theconcentrated load applied to the structure changes, thestrain distribution changes accordingly, and the outputmode of the underground footwork mobile training moni-toring body also changes. For strain distribution monitoring,the main body and functions of the collaborative manage-ment method in this paper are shown in Table 1.

Aiming at the large-scale sports stadium structure, thispaper uses the vibration response of the structure and theactive monitoring method based on the piezoelectric sen-sor array to propose an effective cyclic excitation-sequential sensing scheme to realize real-time screw loos-ening monitoring, that is, to monitor each piezoelectricbody in turn and receive the sensing signals of two adja-cent piezoelectric bodies at the same time, until the endof the cycle. This method is suitable for large-scale struc-tures where the number of screws is large, and there aremany failure modes of the structure, which are easy toalias.

100 piezoelectric sensing nodes are used in the experi-ment. The excitation signal applied to the structure in theexperiment is a sine wave signal with a frequency of120 kHz and a peak value of 5V. 120 kHz is determinedbased on multiple tests, and the structural vibration responseis most sensitive to screw loosening under the excitation ofthis frequency. When the piezoelectric sensor node No. 1excites its connected piezoelectric sensor, the two adjacentnodes 2 and 100 serve as the child nodes in the cluster,and node No. 1 serves as the cluster head to create theSAA body through the DMA body and merges nodes 2and 100. The peak value of the collected structure vibrationsignal; then, nodes 2 to 100 are, respectively, used as thecluster heads of the structure excitation, and SAA sends allthe fusion data to the CCA main body to determine themode of structural screw loosening. Here, SAA adopts adynamic determination method, and the migration destina-tion is updated sequentially according to the needs of thenode.

Figure 4 shows the time-consuming monitoring of wire-less underground footwork mobile training. Similar to strain

0 50 100 150 200 250 300 350 400 450 5002400

3000

3600

4200

4800

5400

6000

6600

7200

7800

Number of genetic iterations

Net

wor

k w

orki

ng li

fe/r

ound

Average lifespanMaximum life

Figure 5: The simulation result output of 5 cluster heads for wireless underground footwork mobile training monitoring in venues of majorsports events.

8 Journal of Sensors

distribution monitoring, the training samples correspondingto the loosening of each screw need to be monitored first.When the periodic monitoring moment is triggered or theuser monitoring command is received, the CCA main bodywill take the data and training samples obtained by theSAA main body to the Okirid. For distance measurement,the calculated minimum value is the corresponding screwloosening position.

5.2. Simulation Experiment of Genetic Algorithm MultitaskAssignment. The coordinates of the base station locationare set to (0, 0), and the other cluster head nodes are distrib-uted in order according to the first quadrant distance of 10meters. The simulation test parameter settings are shownin Table 2.

The working life simulation results of 5 cluster heads inthe wireless underground footwork mobile training andmonitoring network in venues of major sports events areshown in Figure 5. When the network contains 95 clusterheads, the simulation results of the network working lifeare shown in Figure 6. Figure 7 shows the comparison ofthe simulation results of the network working life when thenetwork contains 5~95 cluster heads, using the genetic algo-rithm optimization method in this paper and the direct starnetwork transmission. It can be seen from Figure 7 that themultitask optimization based on the method in this papercan effectively improve the working life compared to thedirect transmission of the star network.

We set the path loss index of different wireless trans-ceivers m = 3 ~ 6, and then, the simulation results of the

0 50 100 150 200 250 300 350 400 450 5001800

2700

3600

4500

Number of genetic iterations

Net

wor

k w

orki

ng li

fe/r

ound

Average lifespanMaximum life

Figure 6: The simulation result output of 95 cluster heads for wireless underground footwork mobile training monitoring in venues ofmajor sports events.

0 10 20 30 40 50 60 70 80 90 1004500

5000

5500

6000

6500

7000

7500

8000

Number of cluster head nodes

Net

wor

k w

orki

ng li

fe/r

ound

Genetic algorithm optimizationDirect star network

Figure 7: Comparison of working life between genetic algorithm and direct star network transmission in this paper.

9Journal of Sensors

network working life under different application environ-ments of venues of major sports events are obtained, asshown in Figure 8. It can be seen from the figure that thegreater the loss of the environment (i.e., harsh environmentapplications), the lower the network working life.

6. Conclusion

As a kind of large-scale space structure building, venues ofmajor sports events are characterized by not only largeinvestment amount, long design life, large scale of construc-tion, etc., but also long-term exposure to factors such as nat-ural environment, aging of their own materials, unevenfoundation settlement, and loads. The impact of the impacton the structure will cause structural damage. When thestructural damage accumulates to a certain level, a serioussudden accident will occur. Therefore, in order to ensurethe safety, durability, and applicability of large space struc-ture buildings, real-time health monitoring of large spacestructure buildings under construction or during service isrequired. This paper takes venues of major sports events asthe monitoring object and studies the application of struc-tural health monitoring in the field of large-span spatialstructures. Through the analysis of structural health moni-toring technology, we combined with ZigBee technologyand GPRS technology; the overall architecture of the healthmonitoring system for venues of major sports events isdesigned. Through further research on the wireless sensornetwork, according to the requirements of health monitor-ing, the overall architecture of the health monitoring systemis determined, and the wireless remote monitoring datatransmission system based on ZigBee technology and GPRStechnology is designed, including hardware design and soft-ware design to realize the underlying data real-time moni-toring. This paper proposes a wireless sensor networkmanagement method based on multiagent cooperation andcombines active and passive structural health monitoring

examples for experimental verification. A network multitaskoptimal allocation strategy based on a genetic algorithm isproposed. However, the stability of communication doesnot consider the complex situation in the venue and the sit-uation of simultaneous multinode collection in the future.Therefore, the signal attenuation test needs to be performedin a normal competition environment, and a bit error ratetest and a system power consumption test are also requiredunder stable conditions.

Data Availability

The data used to support the findings of this study are avail-able from the corresponding author upon request.

Conflicts of Interest

The authors declare that they have no known competingfinancial interests or personal relationships that could haveappeared to influence the work reported in this paper.

Acknowledgments

This work is supported by the “Research on the supervisionsystem of sports events in China” of the National Social Sci-ence Foundation of China (19BTY018).

References

[1] I. P. Guembe, P. Lopez-Iturri, J. J. Astrain et al., “Basketballplayer on-body biophysical and environmental parametermonitoring based on wireless sensor network integration,”IEEE Access, vol. 9, pp. 27051–27066, 2021.

[2] K. Han, G. Xiao, and X. Yang, “An effective stadium monitor-ing control algorithm based on big data in emerging mobilenetworks,” Multimedia Tools and Applications, vol. 78,no. 21, pp. 29971–29987, 2019.

0 10 20 30 40 50 60 70 80 90 100600

1800

3000

4200

5400

6600

Number of cluster head nodes

Net

wor

k w

orki

ng li

fe/r

ound

m=6m=5

m=4m=3

Figure 8: Comparison of simulation results of network working life under different loss environments of venues of major sports events.

10 Journal of Sensors

[3] A. B. Noel, A. Abdaoui, T. Elfouly, M. H. Ahmed, A. Badawy,and M. S. Shehata, “Structural health monitoring using wire-less sensor networks: a comprehensive survey,” IEEE Commu-nications Surveys & Tutorials, vol. 19, no. 3, pp. 1403–1423,2017.

[4] M. O. Benammar, A. B. Abdaoui, H. S. Ahmad, F. A. Touati,and A. B. Kadri, “Real-time indoor air quality monitoringthrough wireless sensor network,” International Journal ofInternet of Things and Web Services, vol. 2, pp. 7–13, 2017.

[5] W. Loga-Księska, J. Sordyl, and A. Ryguła, “Long-term urbantraffic monitoring based on wireless multi-sensor network,”Open Engineering, vol. 10, no. 1, pp. 197–208, 2020.

[6] O. Avci, O. Abdeljaber, S. Kiranyaz, M. Hussein, and D. J.Inman, “Wireless and real-time structural damage detection:a novel decentralized method for wireless sensor networks,”Journal of Sound and Vibration, vol. 424, pp. 158–172, 2018.

[7] H. N. Li, L. Ren, Z. G. Jia, T. H. Yi, and D. S. Li, “State-of-the-art in structural health monitoring of large and complex civilinfrastructures,” Journal of Civil Structural Health Monitoring,vol. 6, no. 1, pp. 3–16, 2016.

[8] H. Asharioun, H. Asadollahi, T. C. Wan, and N. Gharaei, “Asurvey on analytical modeling and mitigation techniques forthe energy hole problem in corona-based wireless sensor net-work,” Wireless Personal Communications, vol. 81, no. 1,pp. 161–187, 2015.

[9] P. Kaewchur, C. Sritong, B. Sriard, and T. Nima, “Energy effi-cient scheduling algorithm to increase the life time of batterypower in wireless sensor networks for structural health build-ing monitoring applications,” Turkish Journal of Computerand Mathematics Education (TURCOMAT), vol. 12, no. 11,pp. 3229–3237, 2021.

[10] S. Lykov, Y. Asakura, and S. Hanaoka, “Positioning in wirelesssensor network for human sensing problem,” TransportationResearch Procedia, vol. 21, pp. 56–64, 2017.

[11] S. Kumar, C. Sethuraman, and K. Srinivas, “Solar poweredautomatic drip irrigation system (SPADIS) using wireless sen-sor network technology,” International research journal ofengineering and technology, vol. 4, no. 7, pp. 722–731, 2017.

[12] S. Hashima and I. I. Mahmoud, “Efficient wireless sensor net-work for radiation detection in nuclear sites,” InternationalJournal of Electronics and Telecommunications, vol. 67, no. 2,pp. 175–180, 2021.

[13] Z. Zhang and Y. Luo, “Restoring method for missing data ofspatial structural stress monitoring based on correlation,”Mechanical Systems and Signal Processing, vol. 91, pp. 266–277, 2017.

[14] P. H. Soares, J. P. Monteiro, H. F. Freitas, R. Zenko Sakiyama,and C. M. Andrade, “Platform for monitoring and analysis ofair quality in environments with large circulation of people,”Environmental Progress & Sustainable Energy, vol. 37, no. 6,pp. 2050–2057, 2018.

[15] Z. Zhen, “Optimal layout of health monitoring sensor in gym-nasium based on improved particle swarm optimization algo-rithm,” Journal of Intelligent & Fuzzy Systems, vol. 35, no. 3,pp. 2813–2819, 2018.

[16] A. Datteo, F. Lucà, and G. Busca, “Statistical pattern recogni-tion approach for long-time monitoring of the G.Meazza sta-dium by means of AR models and PCA,” EngineeringStructures, vol. 153, pp. 317–333, 2017.

[17] J. Teng, W. Lu, R. F. Wen, and T. Zhang, “Instrumentation onstructural health monitoring systems to real world structures,”Smart Structures and Systems, vol. 15, no. 1, pp. 151–167, 2015.

[18] K. S. E. Phala, A. Kumar, and G. P. Hancke, “Air quality mon-itoring system based on ISO/IEC/IEEE 21451 standards,” IEEESensors Journal, vol. 16, no. 12, pp. 5037–5045, 2016.

[19] J. Lloret, M. Garcia, A. Catala, and J. J. P. C. Rodrigues, “Agroup-based wireless body sensors network using energy har-vesting for soccer team monitoring,” International Journal ofSensor Networks, vol. 21, no. 4, pp. 208–225, 2016.

[20] J. P. Amezquita-Sanchez, M. Valtierra-Rodriguez, andH. Adeli, “Wireless smart sensors for monitoring the healthcondition of civil infrastructure,” Scientia Iranica, vol. 25,no. 6, pp. 2913–2925, 2018.

[21] F. Lepage, T. Docquier, V. Lecuire, and J. P. Georges,“TOMAC-WSN: a newWSN efficient protocol for monitoringbig distributed mechanical systems,” IFAC-PapersOnLine,vol. 51, no. 10, pp. 52–57, 2018.

[22] R. Dinesh Jackson Samuel, E. Fenil, G. Manogaran et al., “Realtime violence detection framework for football stadium com-prising of big data analysis and deep learning through bidirec-tional LSTM,” Computer Networks, vol. 151, pp. 191–200,2019.

[23] L. Yang, C. Fu, Y. Li, and L. Su, “Survey and study on intelli-gent monitoring and health management for large civil struc-ture,” International Journal of Intelligent Robotics andApplications, vol. 3, no. 3, pp. 239–254, 2019.

[24] H. Fu, Z. Sharif-Khodaei, and M. H. F. Aliabadi, “An energy-efficient cyber-physical system for wireless on-board aircraftstructural health monitoring,” Mechanical Systems and SignalProcessing, vol. 128, pp. 352–368, 2019.

[25] S. Panchanathan, S. Chakraborty, T. McDaniel et al., “Enrich-ing the fan experience in a smart stadium using internet ofthings technologies,” International Journal of Semantic Com-puting, vol. 11, no. 2, pp. 137–170, 2017.

11Journal of Sensors