Analyzing the Performance of LAR-1 DNDP Algorithms

Background LAR-1DNDP algorithm is a hybrid route discovery algorithm in mobile ad hoc networks (MANETs). It utilizes two route discovery algorithms, the location-aided routing scheme 1 (LAR-1) algorithm and the novel dynamic noise-dependent probabilistic (DNDP) algorithm. LAR-1DNDP was developed to enhance the performance of the hybrid LAR-1 algorithm and (fixed / dynamic) probabilistic algorithm, namely, (LAR-1P) algorithm, in noisy MANETs. Objective: In this paper we develop a hybrid LAR-1 algorithm and the enhanced DNDP algorithm. In the enhanced DNDP algorithm the maximum retransmission probability (p t,pcmin)is calculated as a function of number of first hop neighbors (k) rather than a fixed value. This is why was called k-dependent pt,pcmin. We calculate and compare the performance of the fixed and several k-dependent LAR-1DNDP algorithms using the MANET simulator (MANSim). Effects of nodes densities (n) and nodes speeds (u) on the performance of the LAR-1DNDP are also investigated. Results: The simulation results obtained showed that k-independent LAR-1DNDP algorithm performs better than the k-dependent LAR-1DNDP algorithm. The performance of the LAR-1 DNDP algorithms are improving as n increases as it can provide higher increase in network reach ability (RCH) at less number of retransmissions (RET), i.e., less cost, while node average velocity (u) has insignificant effect on the performance of the LAR-1 DNDP algorithms. Conclusion: This research presented a description and performance evaluation of LAR-1 k-dependent DNDP algorithm. We compare the performance of LAR-1, LAR-1 fixed, LAR-1 DNDP, LAR-1 k-dependent (linear), and LAR-1 k-dependent (parabolic) in terms of reach ability and number of retransmission. We try to increase the reach ability and decrease the number of retransmissions, and to avoid the drawbacks of the broadcast storm problem that is caused by pure flooding algorithm.