Performance Management in Mixed Wireless Networks Using Multi Distortion Measures

We discuss a new approach that uses a modified k-means clustering algorithm for modeling and managing performance in a cellular and ad-hoc mixed wireless network. This paper describes the analytical foundation and provides extensive simulation results to prove that the enhanced clustering scheme can be used to organize nodes in the mixed network so that it's performance is jointly optimized using three different distortion measures: location, traffic, and interference. The paper concludes by showing that the proposed clustering algorithm can be extended by using additional distance measures such that the mixed network is organized and jointly optimized for QoS, spectrum, or interference management.