Approximation of a confidence interval for link distances in Mobile Ad hoc Networks

A Mobile Ad Hoc Network (MANET) is an infrastructure-less network composed of mobile devices. In order to meet the foresight of newly developing MANETs, a method of maintaining a real-time flow despite dynamic topology and random movement of users is required. Mobility prediction is one of the keys to successful design of efficient protocols to find stable and reliable routes in MANETs. An important characteristic of a MANET is the distribution of the link distance between communicating users. In this paper a novel mobility model is developed for users of a MANET wandering in an unlimited area and is used to derive an analytical framework to approximate the communication links distance distribution, providing the horizontal and vertical movements of the nodes follow independent spatial probability distributions. This could be used to minimize the effects of topological changes while balancing the need to support more stable routes. This paper also studies the performance of the approximation method via extensive simulation experiments and statistical tests to verify the validity of the derived distribution in comparison with a simulated movement model with realistic parameters.