Automating Cellular Network Faults Prediction Using Mobile Intelligent Agents

Abstract

This paper presents an approach for prediction of faults in cellular networks using mobile intelligent agents. Cellular networks are uncertain and dynamic in their behaviour hence the application of different artificial intelligent techniques for prediction, detection and identification of network failures, which can lead to more robust handling of unforeseen anomalies within a network environment. In this paper different artificial intelligent techniques are applied in developing platform independent, autonomous and robust agents that can report on any unforeseen anomaly within the infrastructure of a cellular network service provider. The specific design and implementation is done using Java Agent DEvelopment Framework (JADE). The experimental results obtained show 79% prediction accuracy.