Upload
others
View
5
Download
0
Embed Size (px)
Citation preview
NETWORK ANALYSIS AND ROUTING EVALUATION:
THE NARVAL MODULE Foued Melakessou
University of Luxembourg Interdisciplinary Centre for Security, Reliability and Trust (SnT)
Edouard VII ConvenOon Center
SCILABTEC 15-‐16 May 2014
6th InternaOonal Users Conference
Weicker Building
• The University of Luxembourg, founded in 2003, is mul;lingual, interna;onal and strongly focused on research. Its students and researchers have chosen a modern ins;tu;on with a personal atmosphere, close to European ins;tu;ons, interna;onal companies and Luxembourg’s financial centre.
• Interdisciplinary Centre for Security, Reliability and Trust (SnT), created in 2009. – European research centre of excellence and innova;on in Security and Trust
• APSIA: Applied Security and Informa;on Assurance • A&C: Automa;on and Control • NETLAB: Networking Laboratory • RNES: Reliable Networked Energy Systems • SERVAL: Security and Valida;on of Services and Networks • SIGCOM: Signal Processing & Satellite Communica;ons • SVV: SoVware Valida;on and Verifica;on
– High quality and interna;onally aYrac;ve PhD program – Pla[orm for research collabora;on with partners
Partnership Program • Platform for interaction and cooperation with industrial and government partners • Positive impact on the region’s business
BUTLER Project • Design and demonstrate prototype of a comprehensive,
pervasive and effecOve Context-‐Aware informa;on system, which will operate transparently and seamlessly across various scenarios towards a unified Smart Life environment
• Smart Object & Smart Server & Smart Mobile • Domain: Home, Health, Transport, City and Shopping • Internet-‐of-‐Things (IoT): Large number of constrained and
low cost embedded devices – low power consump;on (baYeries) – Limited ROM/RAM (specific Opera;ng Systems: CONTIKI, TinyOS) – Wireless communica;on range (802.15.4) etc.
4
Module DescripOon & Goal
• Analysis of network protocols and algorithms • NARVAL (Network Analysis and Rou;ng eVALua;on)
– Complete soVware environment enabling the understanding of available communica;on algorithms, but also the design of new schemes
– Graph Op;miza;on, Topology, Internet Traffic, Rou;ng, Transmission Protocol, Route Diversity, Mobility, Security, Anonymity, Path Planning, Wireless Sensor Network, etc.
– Target audience: academics, students, engineers and scien;sts
5
NARVAL
Network Analysis and Routing eVALuation
c
Ver 3.0
Network Model • Node Ni(Xi,Yi,Ei) • Link Lj(Hj,Tj,Wj)
• Wj: propaga;on delay, bandwidth, hop count, traffic load, etc.
• Compute the best path between two nodes Ne and Nr in respect with a specific objec;ve func;on to op;mize
6
0
100
200
300
400
500
600
700
800
900
1000
0 100 200 300 400 500 600 700 800 900 1000
R
A
Ne
Mobile Node
Access PointA
Backbone L1
Backbone L2
Backbone L3
Nx
X
Y
Ny
Vx
Vy
Nr
A
A
A
A
NARVAL Requirements
• Scilab >= 5.3.3 • hYp://atoms.scilab.org/toolboxes/NARVAL • New release is under development and will be uploaded soon (Scilab 5.5.0)
7
NARVAL Squeleton 1/3
• NL_F (29 func;ons) – Random generators, – Nodes’ coordinates, – Nodes’ selec;on, – Histogram, Etc.
• NL_G (82 func;ons) – Graph genera;on and modifica;on (Addi;on/dele;on of nodes/edges), – Neighborhood extrac;on, – Visualiza;on tools, – Sta;s;cs, Etc.
• NL_I (30 func;ons) – Connec;on manager, – Packet manager, – Route manager, – Transport protocols (UDP, TCP, MPTCP) and Sliding window manager, Etc.
8
NARVAL Squeleton 2/3
• NL_M (20 func;ons) – MANET/VANET in free/constrained space, – Random direc;on, – Random walk, – Random way point, Etc.
• NL_R (93 func;ons) – Rou;ng algorithms, – AODV, – Spanning tree, BFS and DFS, – Bellman-‐Ford, Dijkstra, Flood, RPL and ARC, Etc.
• NL_S (34 func;ons) – Network security, – AES encryp;on/decryp;on, – RSA encryp;on/decryp;on, – Informa;on slicing, Etc.
9
NARVAL Squeleton 3/3
• NL_T (13 func;ons) – Topology generator, – Waxman algorithm, – Locality model, – Hierarchical model, Etc.
• NL_V (29 func;ons) – Path planning in constrained environment, – Scene defini;on with obstacles, – Computer vision algorithms (dilata;on, erosion, Moravec, etc.) – Visibility graph, Etc.
10
Current Research with NARVAL
• Path extension analysis of P2P communica;on in small 6LoWPAN/RPL networks (MASCOTS’13)
– Longer paths imply larger energy waste – Sta;s;cal analysis of point-‐to-‐point communica;ons (path
hop length) inside random Wireless Sensor Network (WSN) topologies (LR: RPL & LD:Dijksra)
– Impact of the sink loca;on, the network size
• Towards a new way of reliable rou;ng: mul;ple paths over ARCs
11
0.00 2.44 4.89 7.33 9.78 12.22 14.67 17.11 19.56 22.000.00
2.44
4.89
7.33
9.78
12.22
14.67
17.11
19.56
22.00
L D
L R
D
R
A
CD
B
M K
JFL E
N G H I
Rev
RevRev
Rev
Rev
Rev
Rev
1
2
3
4
5
67
8
910
11
1213 14 15
16 17
18
1920212223
24
25
2627
2829
30
31
32 33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
5152
53
54
55
56
57
58
59 60
61
A1
A2 A3
A4
A5
A6
A7
A8
A9
A10
A11
Conclusion & PerspecOve
• NARVAL (Network Analysis and Rou;ng eVALua;on) is a Scilab module enabling the understanding of available communica;on algorithms, but also the design of new schemes in order to evaluate and improve the traffic behavior and distribu;on on network topologies defined by the user.
• hYp://atoms.scilab.org/toolboxes/NARVAL • Future work
– Networking: DNS, DHCP, etc. – New topology generators – Fault tolerance: global repair vs local repair – Data aggrega;on: Wireless Sensor Network – Localiza;on algorithms: Coopera;ve vs Non-‐Coopera;ve – Rou;ng Algorithms: RIP, DYMO, DSR, OLSR (MPR), OSPF, ACO, etc. – Mobility: Gauss-‐Markov, smooth random, reference point group, obstacle, Markovian random walk, simple individual
mobility markovian, generic individual mobility markovian, etc.
12
THANK YOU !
Contact [email protected]