A Cross-layer Neighbour Discovery Algorithm in Ad hoc Networks based on Hexagonal Clustering and GPS

Abstract

Neighbour discovery is an important part of ad hoc networks. In a typical ad hoc routing protocol, each node broadcasts hello messages at a fix time interval to update the latest information of neighbour nodes and tracks neighbour relationship between the nodes. However, the higher density of network nodes, the more collisions happen between the hello messages, the more bandwidths are wasted and the worse network congestion produces. So we design a cross-layer neighbour discovery algorithm, ND_HC, to actualize the neighbour discovery in MAC layer. In this algorithm, the hello messages are produced in MAC layer, and are sent in TDMA method with the help of hexagonal clustering and GPS, which reduce packets collision probability and improve throughput in the network. The NS-2’s simulation results show that compared to the traditional neighbour discovery algorithm in IEEE 802.11, the proposed algorithm improves the efficiency of neighbour discovery obviously.