Competitive Neural Networks for Fault Detection and Diagnosis in 3G Cellular Systems

We propose a new approach to fault detection and diagnosis in third-generation (3G) cellular networks using competitive neural algorithms. For density estimation purposes, a given neural model is trained with data vectors representing normal behavior of a CDMA2000 cellular system. After training, a normality profile is built from the sample distribution of the quantization errors of the training vectors. Then, we find empirical confidence intervals for testing hypotheses of normal/abnormal functioning of the cellular network. The trained network is also used to generate inference rules that identify the causes of the faults. We compare the performance of four neural algorithms and the results suggest that the proposed approaches outperform current methods.