Experimental Study of Pure Flooding Method for Localizing an Anycast Server in Wireless Ad Hoc Networks

This paper presents performance of a flooding based any cast mechanism for locating the nearest server from a group of contents-equivalent servers in real-world wireless ad hoc networks. Any casting can be considered as an advantageous communication paradigm for ad hoc networks where resource efficiency and robustness is of the first importance. A series of experiments are conducted in an outdoor land and carried out by laptop computers running Windows Vista operating system. An application layer any cast based multithreaded program is developed and used for investigations. In experiments, the behavior of some fundamental performance metrics: response ratio, relative network traffic, average response time and duplicate ratio, was investigated with varying number of hops between source and a group of server nodes in the network.

I. INTRODUCTION

An ad hoc mobile network is an autonomous system consisting of mobile hosts that do not rely on the presence of any fixed network infrastructure like: military, law enforcement, and disaster relief operations. In ad hoc networks, nodes are free to move in an arbitrary manner and in cases that mobile nodes cannot reach to the destination, the nodes will relay their messages through other nodes [1]. There are mainly two ways to investigate problems of data transmission in wireless ad hoc networks. First way is simulation modeling [2], [3], [4], [5], which is time and resource effective compared to realworld experiments [6], [7], [8], [9], that require more resources and much more time. However, the behaviour of wireless ad hoc networks in reality is always affected by physical and environmental elements that simulation cannot put all of them into account. In real world experiments however, very accurate and valuable information about characteristics of ad hoc networks is obtained in exchange for larger resources and longer test times.