Implementation and performance analysis of a multi-meshed tree routing protocol for MANETs

As the importance of mobile data connectivity in our daily lives increases, there has been an accompanying increase in the demand for rapidly deployable, short-lived networks which operate without any base infrastructure. This is where Mobile-ad-hoc Networks step in to turn the dream of getting connected anywhere and at any time into a reality and are fast being adopted into civilian applications like vehicular networks and location-based advertising. They have several advantages over infrastructure based wireless networks (like cellular networks) in terms of cost of deployment, ability to use unlicensed bands, etc. However, they are still plagued with several constraints mainly brought about by their dynamic nature. The main concern here is the use of an efficient routing protocol which must be adaptive and able to maintain routes in spite of the changing network connectivity. In this work we investigate a recently-proposed cluster-based routing protocol which utilizes both mesh and tree concepts and aims at minimizing the control overhead required in the setup and maintenance of the network while maintaining robust connectivity. This hybrid protocol, called Multi-Meshed Tree routing protocol, has a unique way of addressing the issues of dynamic adaptation and enhanced connectivity through a hierarchical cluster-based approach along with Virtual Identifiers. Its main advantage over the various other MANET protocols developed is its algorithm simplicity and low message complexity suited for these limited bandwidth, limited power, resource constrained networks. The focus of this work is implementing this protocol (to work with the ns2 simulator), understanding its advantages and conducting some preliminary simulation based comparisons with prevalent Mobile ad-hoc network protocols. We looked at comparison with AODV, DSDV and DSR Protocols and observed MMT to exhibit more resilient connectivity due to redundant routes, particularly for larger sized networks. Further, there was roughly a 1/6th reduction in routing overhead when compared to proactive protocols. We also studied the effects of configurability of cluster size and inherent non-optimality of routes associated with MMT. We foresee the increased use of MANETS in our daily lives, and hope that the advantages of this routing protocol are utilized for future applications.M.S.Includes bibliographical referencesby Rohit Tharej