Analysis the Performance of Vehicles Ad Hoc Network

Abstract

VANET (Vehicular Ad-hoc Network) is an intelligent network technology in wireless communication where the vehicles act as mobile nodes to share data without any central access point for a safety issue. The vehicles send some information about road status and traffic. Due to recent growth of software technology, it becomes necessary to make a step toward using software to check on this network before the implementation. Simulation tools give us a comprehensive study of the network before applying it in a real environment. There are many network simulators which has their own features to distinguish it from other. We should focus on choosing the best one that gives the best results. NS2 is the most common simulator tool, in this project we used NS2 to design the network, which made communication within the network with different routing protocols. We compare many different routing protocols (AODV, AOMDV, DSDV and DSR) based on the various common metrics, throughput, end to end delay, and packet delivery ratio by varying the number of mobile vehicles while applying CBR traffic. The comparison has been done by using simulation tool NS2. To build the network scenario more like a real environment, different mobility models will be considered. Two type of mobility patterns were used in our project. Studying about different routing protocols give us an idea of using the best protocol in different cases under various parameters of a network. We wrote a TCL Script on NS2 to evaluate the behavior of routing protocols that used in VANET network. After executing the simulation in different mobility models, we realized that AODV has the best performance at both and preferable for the large and small environment of network, but it consumes power during transmission. AOMDV gives middle results in all parameters. The large end to end delay appears in DSR protocol. However, DSDV has lower throughput and packet delivery ratio than other routing protocols with a different number of nodes. On the other hand, it acts well in E2E delay when the network size is changed.