N ETWORK A NALYSIS AND R OUTING E VAL UATION : T HE NARVAL M ODULE
Foued Melakessou University of Luxembourg
Interdisciplinary Centre for Security, Reliability and Trust (SnT) Edouard VII ConvenOon Center
SCILABTEC 15-‐16 May 2014
6
thInternaOonal 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.
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 N
i( X
i,Y
i,E
i)
• Link L
j( H
j,T
j,W
j)
• W
j: propaga;on delay, bandwidth, hop count, traffic load, etc.
• Compute the best path between two nodes N
eand N
rin respect with a
specific objec;ve func;on to op;mize
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 Point A
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.
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.
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 22.00
0.00 2.44 4.89 7.33 9.78 12.22 14.67 17.11 19.56 22.00
LD LR
D R
A C D
B
M K
J F
L E
N G H I
Rev
Rev Rev
Rev
Rev
Rev
Rev
1 2 3 4
5
6 7
8 9 10 11
12 13 14 15
16 17
18 19 21 20 23 22
24 25
26 27
28 29
30 31
32 33 34
35 36 37
38 39
40 41 42
43 44
45 46
47 48
49
50
52 51 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.