Implementation of Hybrid Ad-Hoc Routing Protocol S. Swapna Kumar Department of Computer Science & Engineering Sahrdaya College of Engineering & Technology, Trichur, Kerala sswapnakumar@aol.com Dr. M. Nanda Kumar, Dr. Sheeba V.S Department of Electrical Engineering, Department of Electronics & Communication Engineering, Government Engineering College, Trichur, Sreekrishnapuram, Kerala mnkumar.tcr@gmail.com sheebavs@gmail.com Abstract: -- The goal of this application is to implement a hybrid ad-hoc routing protocol, using the 802.11 wireless protocols. Our implementation enables communication between several wireless stations, on a dynamic network without using any infrastructure, i.e. using peer-to-peer mode, rather than Access Points. Two distant units can communicate even when there is no direct connection between them. We have implemented the Direct Sequence Distance Vector (DSDV) algorithm, which is a pro-active table driven algorithm in Wireless Network. The routing in each station is executed according to local routing table. The tables are continually maintained and updated. We developed the application in Java, which has inherent support for network operations. Thus, it is platform independent, and can run with various OS and wireless cards. In order to demonstrate the operation of the algorithm we wrote a Unicode Short Message Text (USMT) application that uses the routing protocol services as a sub-layer. The USMT application enables sending of text messages from any unit to any other unit in the network. It also graphically presents the local routing tables information. Keywords: AODV, DSDV, MANET, Wireless Ad-hoc Network, Test-bed for Ad-hoc implementation I. INTRODUCTION Simulation is an important tool in the development of MANET as well as in certain dynamic applications of wireless sensor networks; it provides an excellent environment to experiment and verify routing protocol correctness. However, simulation does not guarantee that the protocol works in practice, because simulators contain assumptions and simplified models that may not actually reflect real network operation. After a protocol is thoroughly tested in simulation, an implementation is the logical next step. A working implementation is necessary to validate that the routing protocol specification performs under real conditions. Otherwise, assumptions made by the protocol design cannot be verified as correct. Additionally, an implementation can be used to perform test bed and field tests. Eventually it can be used in a deployed system. Creating a working implementation of an ad-hoc routing protocol is non- trivial and more difficult than developing a simulation. In simulation, the developer controls the whole system, which is in effect only a single component. An implementation, on the other hand, needs to interoperate with a large, complex system. Some components of this system are the operating system, sockets, and network interfaces. Additional implementation problems surface because current operating systems are not built to support ad-hoc routing protocols. A number of required events are unsupported; support for these events must be added. Because these events encompass many system components, the components and their interactions must also be explored. For these reasons it takes significantly more effort to create an ad hoc routing protocol implementation than a simulation II. PROBLEM STATEMENT To date, the majority of ad-hoc routing protocol research has been done using simulation only. One of the most motivating reasons to use simulation is the difficulty of creating a real implementation. In a simulator, the code is contained within a single logical component, which is clearly defined and accessible. On the other hand, creating an implementation requires use of a system with many components, including many that have little or no documentation. The implementation developer must understand not only the routing protocol, but all the system components and their complex interactions. Further, since ad hoc routing protocols are significantly different from traditional routing protocols, a new set of features must be introduced to support the routing protocol. In this paper we describe the event triggers required for AODV operation [1], the design possibilities and the decisions for our Ad hoc On- demand Distance Vector (AODV) routing protocol implementation, AODV-DSDV. This paper is meant to aid researchers in developing their own on-demand ad hoc routing protocols and assist users in determining the implementation design that best fits their needs. III. MOTIVATION Till date, the majority of ad-hoc routing protocol research has been done using simulation only. One of the most motivating reasons to use simulation is the difficulty of creating a real implementation. In a simulator, the code is contained within a single logical component, which is clearly defined and accessible. On the other hand, creating an implementation requires use of a system with many components, including many that have little or no documentation. The implementation developer must understand not only the routing protocol, but all the system components and their 2010 International Conference on Advances in Recent Technologies in Communication and Computing 978-0-7695-4201-0/10 $26.00 © 2010 IEEE DOI 10.1109/ARTCom.2010.14 151