Realistic Mobility Models for Vehicular Ad hoc Network (VANET) Simulations

Kun-chan Lan and Chien-Ming Chou

National Cheng Kung University

Speaker ：潘逸峻

Adviser ： 許子衡 老師

Outline

• Introduction– Important simulation parameters– Simulation tools (MOVE)– Experimental Background

• Effects of details of mobility models– Existence of traffic lights– Driver route choice– Overtaking behavior

• Conclusion and Future work

Introduction

VANET

Enable quick and cost-efficient distribution of data for the passengers.

Simulation tools

Test and evaluate network protocols. EX ： ns-2 , OPNET and Qualnet.

Important simulation parameters

Node mobility model consist a realistic topological

map: different densities of roads and

different categories of streets with various speed limits.

Deceleration and acceleration model.

Turning model : decide a turning direction at the intersection.

Mobility model generator for Vehicular networks(MOVE)

Rapidly generate realistic mobility models for VANET simulations.

Built on top of an open source micro-traffic simulator SUMO.

Map Editor

Create the road topology. Three ways to create the road map

1. Created by the user.

2. Generated automatically.

3. Imported from existing real world maps.

Vehicle Movement Editor

Specify the trips of vehicles and the route . Three methods to define the vehicle

movements

1. Created by the user.

2. Generated automatically.

3. Specified based on a bus time table to

simulate the movements of public transport.

Experimental Background

• Nodes location, density, and direction etc.

affect VANET performance directly.

• Mobility models generated : MOVE

• Road topology generated : TIGER database

• Node : 802.11 MAC operating at 2Mbps

• Transmission range : 250m

Effects of details of mobility models

• Three case studies

– Traffic lights.

– Driver route choice.

– Car overtaking.

• Number of nodes : 300.

• Simulation time lasts : 1000 seconds.

• Roads created : two lanes.

Existence of traffic lights

• Clustering effect

– a traffic light have a higher node density.

• improve the network connectivity

– higher chance for packet collision

• The distance between two traffic lights

• A larger number of neighboring nodes typically suggests a better network connectivity

• Packet delivery ratio decrease when there are

more traffic sources.

• Two adjacent clusters which significant

degrades the network performance.

• traffic light cycle can also have a significant

impact on the network performance.

Driver route choice

• Different choices of route directions can

significantly change the simulation results.

Overtaking behavior

• when overtaking behavior is not allowed, it usually results in a chain-like topology and a shorter and uniform inter-vehicle distance

Conclusion and Future work

• In this paper, we show that the details of a mobility model can have a significant impact on the simulation results

• Build an interface to tightly integrate SUMO and ns-2 , vehicle state information (such as location, speed, direction, etc.) can be fed into ns-2 in real time.