A location-based routing protocol for mobile wireless ad-hoc networks

Abstract

Mobile Ad hoe networks (MANETs) consist of a number of nodes all equipped with a wireless communication device that allow communication between them over a wireless medium. In this work, a Location Service Module (LSM) was developed and used by a location-based routing protocol (LBRP) to route packets on MANETs. Location-based routing protocols base their routing decisions on their current, the neighbours', and destination's location. Thus, nodes require a way to acquire these positions. The LSM is used for this purpose; where it implements a beacon based some-for-all location service with some of the nodes on the network, referred to as Location Service Nodes (LSN), learning the location of all the other nodes on the network and supplying the location information of nodes that are not within the requesting nodes' range. The LSM module resides on each node and periodically broadcasts the node's location and velocity, so that neighbour nodes store these details in their location table. This is later used by the LSM to provide the routing protocol with the next hop details towards destination. The neighbour nodes' velocity allows the reduction of the location beaconing frequency by predicting the location of neighbour nodes. The LSNs periodically exchange their location tables (LTs) and check whether they can handover to other LSNs to reduce overheads in propagating their LT. The LBRP implemented uses a greedy forwarding approach, where packets are forwarded to the neighbour node which is closest to the destination than the current node. The LBRP triggers the LSM to learn the next hop towards the destination node and the transmission power required when sending packets. Furthermore, it uses a back tracking list which is added to every packet so that it can recover from local minima and avoid looping in doing so. In the performed simulations, the proposed solution performed well in both dense and non dense networks. When gauged in dense environments having some of the nodes sending packets, the LBRP, managed to deliver on average 99.7% of the packets indicating that the LSM is scalable. On the other hand, it manages to deliver 98.1 % of the packets when all the nodes on the network are sending messages.