SHORT PAPER International Journal of Recent Trends in Engineering, Vol. 1, No. 1, May 2009 463 Timing Analysis and Repeatability Issues of Mobile Ad-Hoc Networking Application traffics in Large Scale Scenarios V.R. Sarma Dhulipala 1 , RM.Chandrasekaran 2 , and R.Prabakaran 1 1 Anna University Tiruchirappalli/Center for Convergence of Technologies, Tiruchirappalli, India 1 Email: {dvrsarma,hiprabakaran}@gmail.com 2 Anna University Tiruchirappalli, Tiruchirappalli, India 2 Email: aurmc@sify.com Abstract- Present day mobile ad hoc networks grow in size and the associated complexity warrant the need to explore the suitability of the various application traffics and their repeatability during their execution. This paper presents a study on the timing analysis and their repeatability for common MANET (mobile ad hoc network) application traffics in large-scale scenarios. The application traffics analyzed in this study include Constant Bit Rate (CBR), File Transfer Protocol (FTP), Telecommunication network (Telnet) and Variable Bit Rate (VBR). We studied the execution and real times by conducting simulation experiments with multiple mobile node scenarios for the above said application traffics. We also studied the run time variations (repeatability issues) for various interactive runs. The results were graphically plotted and explained. Index Terms: Ad hoc networks, Repeatability, MANET, CBR, FTP, TELNET and VBR, fault tolerance, wireless senor node I. INTRODUCTION Mobile Ad hoc Networks (MANETs) offer communication over a shared wireless channel and are extensively employed, however without any pre-existing infrastructure [1]. Communications must be set up and maintained on the fly over mostly by wireless links. Each node of an ad hoc network can both route and forward data [2]. The exploding demand for computing and communication on the move has led to reliance for ad hoc networks. Although substantial attempts have been made on research towards the design and development of ad hoc APPLICATION TRAFFIC, there is relatively little understanding of their behaviour in terms of the performance [3, 4] by comparing execution and real times as the system is scaled up. First, it is unclear whether any existing ad hoc application traffics can be used to successfully perform within the given terrain dimensions which will be necessary in large scale wireless networks. Second, it is also ambiguous that which of the preceding traffics is likely to perform better as the size of the network grows [5]. This lack of understanding arises from the fact that previous simulation tools might not be sufficient enough to facilitate the analysis of large scale networks without compromising the fidelity, and thus the accuracy, of the ad hoc network models. In general, the definition of scalability is that the application traffic under use should be able to provide low execution time as well as a lower percentage deviation relative to the network scenario. Furthermore, it also should produce consistency from run to run and the application traffic itself could use minimal routing control overhead and exhibit high packet delivery ratio (which correlates to high application throughput) [5]. Finally, for application traffic to be considered completely scalable, the aforementioned characteristics are expected to hold true regardless of the network size, node density, mobility and number of sources and destinations and degrade gracefully if the network configuration has insufficient resources. It is also important to satisfy the repeatability of a simulation study. In this paper, we analyse the scalability of selected application traffics including Constant Bit Rate (CBR), File Transfer Protocol (FTP), TELecommunications NETwrok (TELNET), and Variable Bit Rate (VBR). We have used a more generalised Ad hoc On-Demand Distance Vector (AODV) routing protocol for a simple and large scale Client – Server communication in which communication is handled through Hyper Text Transfer Protocol (HTTP) through simulation in the application layer. We detail our general methodology for ascertaining the scalability of ad hoc protocol and discuss the results from the present studies. II.GENERAL METHODOLOGY From the existing tools ((like NS 2, GloMoSim, etc.,) Fig 1: Mobile Ad hoc Network © 2009 ACADEMY PUBLISHER