A Method of Opportunity Prediction in Mobile Ad Hoc Network

Abstract

The mobility of nodes in mobile ad hoc network (MANET) makes great changes in network topology. And limited knowledge of the future encounter opportunity leads to a blind and unpredictable packet forwarding behavior in routing decisions. Finding the regularity in topology changes and makes efficient opportunity prediction is the key to routing in mobile ad hoc network. To this end, this thesis proposed KROP, a kernel regression opportunity prediction method in MANET. In KROP, we first extract Adamic-Adar metric and contact frequency metric to form features of node pairs, and use these features to capture the evolution of the local network topology over time. Then, we use kernel regression estimation method to model the historical evolution of the topology and output the probability of a future encounter. According to the given sequence of the probability, we make opportunity prediction of the future. In the comparsion experiments on world datasets, we eventually proved KROP outperforms on prediction accuracy.