Topology Control in MANETs Using the Bayesian Pursuit Algorithm

摘要

A way of optimizing energy consumption in mobile ad hoc networks is topology control. Various topology control algorithms have been proposed, but a few of them have used meta-heuristic methods such as genetic algorithms, neural networks, and learning automata. This paper presents a new algorithm based on the Bayesian pursuit (BPST) algorithm. The proposed algorithm predicts node mobility parameters by learning from the environment and uses these parameters to predict link availability duration. The algorithm, then, finds the paths with maximum lifetime by removing the links with a short lifetime using the local Dijkstra algorithm. In this way, an efficient and reliable topology is constructed. The paper also provides the proof of the convergence of the propounded BPST algorithm. The simulation results show that this algorithm can efficiently reduce energy consumption and improve throughput and end-to-end delay as two parameters of the quality of service.