Cellular network coverage optimization through the application of self-organizing neural networks
Carl James Debono, Julian K. Buhagiar · 2006
Site cluster coverage optimization is crucial in improving the quality of service offered by a cellular system. An optimization technique using a two-layered self-organizing neural network is presented. The kernel maps the deterministic patterns obtained through actual traffic data and produces a performance relationship with regard to cluster size. Optimum configuration for a given cellular cluster can thus be determined as the kernel changes hardware parameters with the aim of enhancing coverage. The algorithm's resulting tuning parameters can later be applied to the cellular network's equipment. Simulation results have demonstrated the efficacy of the kernel in improving cellular network coverage.