On the Use of Smart Ants for Efficient Routing in Wireless Mesh Networks

Abstract

Routing in wireless mesh networks (WMNs) has been an active area of research for the last several years. In this paper, we address the problem of packet routing for efficient data forwarding in wireless mesh networks (WMNs) with the help of smart ants acting as intelligent agents. The aim of this paper is to study the use of such biologically inspired agents to effectively route the packets in WMNs. In particular, we propose AntMesh, a distributed interference-aware data forwarding algorithm which enables the use of smart ants to probabilistically and concurrently perform the routing and data forwarding in order to stochastically solve a dynamic network routing problem. AntMesh belongs to the class of routing algorithms inspired by the behaviour of real ants which are known to find a shortest path between their nest and a food source. In addition, AntMesh has the capability to effectively utilize the space/channel diversity typically common in multi radio WMNs and to discover high throughput paths with less interflow and intra-flow interference while conventional wireless network routing protocols fail to do so. We implement our smart ant-based routing algorithm in ns-2 and carry out extensive evaluation. We demonstrate the stability of AntMesh in terms of how quickly it adapts itself to the changing dynamics or load on the network. We tune the parameters of AntMesh algorithm to study the effect on its performance in terms of the routing load and end-to-end delay and have tested its performance under various network scenarios particularly fixed nodes mesh networks and also on mobile WMN scenarios. The results obtained show AntMesh's advantages that make it a valuable candidate to operate in mesh networks.