A Novel Dynamic Noise-Dependent Probabilistic Algorithm for Route Discovery in MANETs

In mobile ad hoc networks (MANETs), broadcasting is widely used in route discovery and other network services. The most widely used broadcasting algorithm is simple flooding, which aggravates a high number of redundant packet retransmissions, causing contention and collisions. Proper use of dynamic probabilistic algorithm significantly reduces the number of retransmissions, which reduces the chance of contention and collisions. In current dynamic probabilistic algorithm, the retransmission probability (pt) is formulated as a linear/non-linear function of a single variable, the number of first-hop neighbors (k). However, such algorithm is suffers in the presence of noise due to increasing packet-loss. In this paper, the authors propose a new dynamic probabilistic algorithm in which pt is determined locally by the retransmitting nodes considering both k and the noise-level. This algorithm is referred to as the dynamic noise-dependent probabilistic (DNDP) algorithm. The performance of the DNDP algorithm is evaluated through simulations using the MANET simulator (MANSim). The simulation results show that the DNDP algorithm presents higher network reachability than the dynamic probabilistic algorithm at a reasonable increase in the number of retransmissions for a wide range of noise-level. The effects of nodes densities and nodes speeds on the performance of the DNDP algorithm are also investigated.