Performance evaluation of MANETs virtual backbone formation algorithms

Abstract

Mobile Ad hoc networks (MANETs) are gaining increased interest as the technology that potentially will make mobile computing a tangible reality. The self-control, selforganization, topology dynamism, and bandwidth limitation of the wireless communication channel make MANETs' implementation a challenging task. The Connected Dominating Set (CDS), a.k.a. virtual backbone or spine, has been proposed to facilitate routing, broadcasting, and establishing a dynamic infrastructure for distributed location databases in MANETs. Minimizing the CDS cardinality simplifies the MANET's abstracted topology and allows for using shorter routes. Since finding the minimum size CDS (MCDS) is NP-complete, approximation algorithms and heuristics are used to tackle this problem. It has been reported that localized CDS creation algorithms run fast and generate light signaling overhead; examples of these algorithms are Wu and Li algorithm and its Stojmenovic variant, the MPR algorithm, and Alzoubi algorithm. Some theoretical performance analysis of these algorithms is presented in the literature; however, this analysis has not been validated across any physical or at least simulation-based measures. Moreover, the presented theoretical performance analysis has overlooked the cost of maintaining the CDS in the presence of topology changes. In this paper, a simulation-based evaluation of the performance of these algorithms is conducted using the ns-2 network simulator. Moreover, the effects of mobility rates and network size on the performance of each algorithm is investigated.