Efficient Routing Protocols for Vehicular Communication Networks Katsaros Konstantinos PhD Student Supervisor Dr M Dianati Cosupervisor Prof R Tafazolli T ransfer Presentation ID: 557037
Download Presentation The PPT/PDF document "Reliable and" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Reliable and
Efficient Routing Protocols for Vehicular Communication Networks
Katsaros Konstantinos PhD Student
Supervisor: Dr. M. Dianati Co-supervisor: Prof. R. Tafazolli
T
ransfer PresentationSlide2
Outline
Introduction
Scope, Objectives, ChallengesRouting in VANETsTaxonomy, Forwarding techniques, Recovery strategies
, Cross-layeringAchievements so farProposed CLWPR (System model, design characteristics)
Performance evaluation
Future plan
Konstantinos Katsaros
2Slide3
Scope
Intelligent Transportation Systems (ITS)Application of Information and Communication Technologies for future transport systems
In order to:Improve safety and traffic managementProvide infotainment services. Vehicular Communications is an important part of ITS.
Cellular (3G, LTE) and Dedicated Short Range Communications (IEEE 802.11p / WAVE)
Konstantinos Katsaros
3Slide4
VANETs: Challenges & Opportunities
Are a category of Mobile Ad-hoc Networks (MANETs) with specific characteristics:Less strict energy and computational constraints
Highly dynamicPredictable mobility patterns
High density of nodes
Konstantinos Katsaros
4Slide5
Objectives of this work
To design reliable and efficient routing protocols by exploiting:Position and mobility information in order to increase efficiencyPHY and MAC information in order to increase reliability
To design a Location Service that can provide position information for the routing protocols
Konstantinos Katsaros
5Slide6
BACKGROUND
Overview of routing and forwarding protocols for MANETs and VANETs
Konstantinos Katsaros
6Slide7
Routing Taxonomy
Advantages
Disadvantages
Routing Protocols
for VANETs
Topology Based
Proactive
Do not flood entire network
Fast
path selection
Overhead to maintain tables
Reactive
Do not maintain routing
tables
Initial
delay for route discovery
Flood a route request
Hybrid
Combination of proactive
and reactive in different
operation stages
Hierarchical
Exploit clusters with similar characteristics
Overhead to maintain clusters
Flooding
Low
complexity, high data reception
Flood entire
network
Position Based
With
out Navigation
Rely on local
information only
Need a
location service (LS), more prone to local maximum problem
With NavigationExploit mobility of nodes, less prone to local maximumNeed a LS, increased overhead due to enhanced beaconing
Konstantinos Katsaros
7Slide8
Position-based Forwarding without Navigation
Konstantinos Katsaros
8
S
3
5
1
2
D
4
6
7
8
Greedy Forwarding
Most Forward in Radius
Nearest Forwarding Progress
Compass
Random Positive ProgressSlide9
Local Maximum Problem & Recovery Techniques
Konstantinos Katsaros
9
S
D
Recovery
strategies
:
Drop packet
Enhanced Greedy (random retransmission once)
Carry-n-Forward
Coloring
Left hand rule
Perimeter routingSlide10
Position-based Forwarding with Navigation
Konstantinos Katsaros
10
“Anchor” points at junctions with coordinator nodes
Enhanced beacon messages with velocity/heading
Position prediction policy (dead reckoning)
Estimation of link lifetime
Vehicle traffic information (max velocity, traffic density)
Recovery From Local
Maximum
Re-route
using different anchor points (with or without deletion)Slide11
Cross-Layer Optimization of Routing Protocols
Network layer with PHY and MAC: Use channel/link quality information for routing
decisionNetwork layer with Transport and Application
: Provide different levels of priorities on packets
Konstantinos Katsaros
11Slide12
CROSS-LAYER POSITION BASED ROUTING (CLWPR)
Proposed routing protocol: system model and design characteristics
Konstantinos Katsaros
12Slide13
System Model
Important Assumptions:Position and navigation information are available (e.g., using GPS)Nodes are equipped with the IEEE 802.11p based communication facility
Konstantinos Katsaros
13Slide14
Main Features of CLWPR
Unicast, multi-hop, cross-layer, opportunistic routingNeighbor discovery based on periodic 1-hop “HELLO” messages“HELLO” message content: position, velocity, heading, road id, node utilization, MAC information, number of cached packets
total size 52
bytesUse of position prediction and “curvemetric” distanceUse of SNIR information from “HELLO” messagesEmploy carry-n-forward strategy for local-maximum
Combine metrics in a weighting function used for forwarding decisions
Konstantinos Katsaros
14Slide15
Weighting Function for Next Hop Selection
The node with the
least
weight will be selected
Currently f
i
weights are fixed – open issue to optimize them or use adaptive values
Konstantinos Katsaros
15Slide16
PERFORMANCE EVALUATION
Simulation setup, initial results, performance analysis and comparison
Konstantinos Katsaros
16Slide17
Simulations Setup
Performance metricsPacket Deliver Ratio (PDR), End-to-End Delay, network overhead.
Use ns-3 for simulations5x5 grid network, 200 and 100 vehicles scenarios10 concurrent vehicle-to-vehicle connections UDP
packets (512 Bytes) with 2 sec intervalIEEE 802.11p, 3Mbps, RTS/CTS enabledTwo-Ray-Ground model
Konstantinos Katsaros
17Slide18
Comparison with GPSR
Konstantinos Katsaros
18
Increased PDR
Reduced end-to-end delay
Increased overhead due to larger HELLO messagesSlide19
Impact of HELLO interval and prediction
Konstantinos Katsaros
19
Prediction improves PDR
More frequent HELLO increases PDR
Network overhead could be reduced by increasing HELLO interval for the same PDR threshold.Slide20
Influence of navigation
Konstantinos Katsaros
20
Navigation improves PDR
Increasing weight of navigation information has positive effect in higher vehicle speedsSlide21
Influence of SNIR
Konstantinos Katsaros
21
SNIR information reduces end-to-end delay
Due to propagation model used, not big improvements
Expect more when shadowing is includedSlide22
Influence of
Carry-n-Forward
Konstantinos Katsaros
22
Increased PDR with time of caching
Increased end-to-end delay with time of cachingSlide23
FUTURE WORK
CWPR optimization, proposed location service, impact assessment and security issues
Konstantinos Katsaros
23Slide24
Future Work (1)
CLWPR OptimizationUse realistic propagation modelOptimize all weighting parametersLocation Service (a)
RSUs as distributed databaseCo-operation between nodesReduce number and latency of queries
Konstantinos Katsaros
24Slide25
Future Work (2)
Location Service (b) – heterogeneous networkUse of UMTS technologies for control and signaling to provide location service
Impact AssessmentAsses impact of ITS applications on network reliabilitySecurity IssuesAnalyze potential threats on reliability of vehicular networks, specially for Location services
Konstantinos Katsaros
25Slide26
Work Plan
Konstantinos Katsaros
26Slide27
Publications
Current:K. Katsaros, et al. “CLWPR - A novel cross-layer optimized position based routing protocol for VANETs
", in IEEE Vehicular Networking Conference, pp. 200-207, 2011K. Katsaros, et al. “
Application of Vehicular Communications for Improving the Efficiency of Traffic in Urban Areas", accepted in Wireless Communications and Mobile Computing, 2011.K. Katsaros, et al. ”Performance Analysis of a Green Light Optimized Speed Advisory (GLOSA) application using an integrated cooperative ITS simulation platform
", in Proceedings of
IEEE International Wireless Communications and Mobile Computing Conference (IWCMC)
,
pp. 918 - 923, 2011Planned:
Survey Paper on routing protocols for VANETsConf. paper @ NS-3 Workshop in SIMUTools 2012, regarding the architecture and implementation (Nov. ‘11)Journal article @ JSAC on Vehicular Communications extending CLWPR paper (Feb. ‘12)
Konstantinos Katsaros
27Slide28
Questions
Konstantinos Katsaros
28
Email:
K.Katsaros@surrey.ac.uk
www:
info.ee.surrey.ac.uk/Personal/K.Katsaros/Slide29
Current work
Propagation Loss Model for urban environment, initial results
Konstantinos Katsaros
29Slide30
Winner B1 model for urban V2V
Konstantinos Katsaros
30
[1]
IST-WINNER D1.1.2 P. Kyösti, et al., "WINNER II Channel Models", September 2007
.
Available
at: https://www.ist-winner.org/WINNER2-Deliverables/D1.1.2v1.1.pdf
Use propagation models from [1] taking into account buildings and shadowing with LOS and NLOS componentsSlide31
TwoRayGround Vs. Winner in network graph / connections
Konstantinos Katsaros
31Slide32
TwoRayGround Vs. Winner in PDR
Konstantinos Katsaros
32Slide33
Cross-Layer Designs (1)
Network layer with PHY and MAC: Use channel/link quality information for routing decision
Link Residual TimeSNR info for MuiltiPoint Relay selectionMAC layer position information for prediction
MAC retransmissionsDeReHQ [1]: Delay, Reliability and Hop countPROMPT [2]: Delay aware routing and robust MACMAC collaboration for heterogeneous networks
[1]
Z.
Niu
, W. Yao, Q. Ni, and Y. Song, “Study on QoS Support in 802.11e-based Multi-hop Vehicular Wireless Ad Hoc Networks,” in IEEE
International Conference on Networking, Sensing and Control, pp. 705 –710, 2007.
[2] B. Jarupan and E. Ekici, “PROMPT: A cross-layer position-based communication protocol for delay-aware vehicular access networks,” Ad Hoc Networks, vol
. 8, pp. 489–505, July 2010.
Konstantinos Katsaros
33Slide34
Cross-Layer Designs (2)
Network layer with transport and Application: Provide different levels of priorities on packetsVTP (Vehicular Transport Protocol)
Optimization of TCP and GPSR with vehicle mobility (adaptive beacon interval)Network layer with multiple layersJoint MAC, Network and Transport [1]
Konstantinos Katsaros
34
[1]
L. Zhou, B.
Zheng
, B. Geller, a. Wei, S.
Xu
, and Y. Li, “Cross-layer rate control, medium access control and routing design in cooperative
VANET”, Computer
Communications, vol. 31, pp. 2870–2882, July 2008Slide35
Location Services
Flooding based: All nodes host itProactive: DREAM
Reactive: LAR, MALM (mobility assisted)Rendezvous based: Some nodes host itQuorum: divide node set into two subsets (update and query)
Hashing (according to node ID or location): define server nodes using a hash functionRLSMP (Region-based Location Service Management Protocol) and MG-LSM (Mobile Group Location Service Management) designed for VANETs utilizing mobility information
Konstantinos Katsaros
35