Reproducing and Extending Real Testbed Evaluation of GeoNetworking Implementation in Simulated Networks

Abstract

International audienceVehicular Ad-hoc Network (VANET) is a type of Mobile Ad-hoc Network (MANET) which is specialized for vehicle communication. GeoNetworking is a new standardized network layer protocol for VANET which employs geolocation based routing. However, conducting large scale experiments in GeoNetworking softwares is extremely difficult, since it requires many extra factors such as vehicles, stuff, place, terrain, etc. In this paper, we propose a method to reproduce realistic results in simulation with the same software implementation. The key idea of the method is to calibrate simulator with the results of real world testbed experiments. After the simulator was calibrated, some extended experiments were carried out. Through these experiments, the fundamental functions of the GeoNetworking implementation (BTP, Greedy Forwarding, etc.) are verified, while an issue in algorithm was discovered and analyzed

DUPE: Duplicated Unicast Packet Encapsulation in Position-based Routing VANET

“…In section IV, we discuss the additional requirements and constraints of the routing protocol, then propose a routing encapsulation layer called Duplicated Unicast Packet Encapsulation (DUPE) to solve the packet loss issue. In section V, several experiments of DUPE is conducted by the evaluation method of Combined Realistic Evaluation Workflow (CREW) , which is proposed in [3]. All of them show a fairly promising result: we efficiently eliminated the packet loss, with a reasonable overhead.…”

“…In this paper, we conduct the experiments within the Combined Realistic Evaluation Workflow (CREW) [3]. The workflow contains three steps: experimentation, reproduction and extension.…”