LOCAL ADAPTIVE TOPOLOGY CONTROL IN MOBILE AD HOC NETWORKS

Abstract

One of the main challenges in mobile ad hoc networks is to establish a relationship between the model of node mobility and the routing protocols and topology control used to ensure that the network holds connectivity and supports efficient communication, valuing always the lowest possible power consumption . However, several recent studies analyze these three aspects in isolation, assessing only: a) the performance of a mobility model without mentioning any protocols, b) the performance of a routing protocol associated with a mobility model, but without using a protocol for topology control, and c) the performance of a topology control protocol associated with a model of mobility, regardless of a routing protocol.Given a mobility model and a routing protocol, the goal of this project is to explore and propose techniques for topology control protocols that improve network efficiency, both in energy consumption and transmission capacity. These techniques may be based on algorithms that work with local adaptations, where the graph changes are made to ensure some structural properties. Another approach to be investigated is based on the use of probabilistic methods using Markov models, where modifying the protocol parameters (such as transmission range at every node) can be performed using a reinforcement learning algorithm.