Efficient node distribution techniques in mobile ad hoc networks using game theory

Self deployment of nodes in mobile ad hoc networks (MANETs) is a challenging task due to the characteristics of a MANET such as dynamically changing topology, lack of centralized authority, decentralized architecture and heterogeneous nodes. In military applications where the deployed nodes, due to their limited communication ranges and hostile environment, may act selfishly with conflicting individual interests among their neighbors, game-theoretic approaches become relevant. Using our distributed game (NSPG-G1) for MANET nodes to position themselves in an unknown geographical terrain to maximize the area coverage, we show that, combined with a distributed genetic algorithm (GA) to determine the next best location to move, NSPG-G1 can provide a near uniform node spreading. In this distributed and scalable game participated by the neighboring nodes autonomously, the decisions about node movements are solely based on localized data about the neighboring nodes while requiring a limited synchronization among a player's closest neighbors.