International Journal of Innovative Research and Development

Abstract

A recent upcoming smart traffic is Vehicular ad hoc network (VANET) is also considered as a sub-set of mobile ad hoc network (MANET). It provides wireless ad-hoc communication in between vehicles and vehicle to roadside equipments. Based on this technology traffic  network is classified into two types 1. vehicle to vehicle interaction, 2. vehicle to infrastructure interaction. The objective of VANET is to provide safe, secure and automated traffic system. For this automation of traffic technique is different types of routing protocols have been developed. But routing protocols of MANET are not directly applicable to VANET.

VANET is designed on IEEE 802.11b wireless standard. This helps to communicate vehicle to vehicle and vehicle to trafic communications. According to Federal Communications Commission (FCC) suggests for VANET frequency spectrum of 75 MHz in the range of 5.850 GHz to 5.925 GHz. It communicate  from one vehicle (source) to another vehicle (destination) through different vehicles (intermediate nodes). A numbers of different routing protocols for communication, ie multimedia data, text data etc. from one vehicle (node) to another vehicle are existing. The Ad hoc On-Demand Distance Vector (AODV)[1] routing algorithm is one of the popular routing  protocols for ad-hoc mobile networks. AODV is used for both unicast and multicast routing. Earlier we have modified AODV with stack and dqueue, where we have find a considerable amount of betterment of result with respect of AODV.  In this paper, we propose and implement in the NCTUns-6.0 simulator neural network based Modified AODV on dqAODV(dqueue implemented AODV)[2] and sAODV (stack implemented AODV)[3] routing protocol considering Power, TTL, Node distance and Payload parameter to  find the optimal route from the source station (vehicle) to the destination station in VANET com-munications.  Further we compare both neural network optimized dqAODV (dqueue implemented AODV) and sAODV (stack implemented AODV) performance on a signal fading model (Rayleigh). This gives us a better result in ai-sAODV(Neural network optimized Stack implemented AODV) compared to queue implemented AODV (ai-dqAODV).

Keywords

Full Text