Bayesian Learning of Neural Networks for Mobile User Position Prediction

Abstract

In this paper, a novel technique for location prediction of mobile users has been proposed, and a paging technique based on it is developed. Mobile users are creatures of habits. They tend to repeat their behaviors. Hence, neural networks with its learning and generalization ability may act as a suitable tool to predict the location of a mobile user provided it is trained appropriately by the personal mobility profile. For prediction, a novel hybrid Bayesian neural network model for predicting locations on Cellular Networks (can also be extended to other wireless networks such as Wi-Fi and WiMAX) is suggested. We investigate its different parallel implementation techniques on mobile devices, and compare its performance to many standard neural network techniques such as: Back-propagation, Elman, Resilient, Levenberg-Marqudat, and One-Step Secant models. This approach is free from all unrealistic assumptions about the movement of the users. It is applicable to any arbitrary cell architecture. It attempts to reduce the total location management cost and paging delay. In general, it enhances mobility management in wireless networks (in location management and hand-off management). In our experiments, we compare results of the proposed Bayesian Neural Network with 5 standard neural network techniques in predicting next location. Bayesian learning for Neural Networks predicts location better than standard neural network techniques since it uses well founded probability model to represent uncertainty about the relationship being learned. The result of Bayesian training is a posterior distribution over network weights.