Performance modelling and bandwidth management of WiMAX systems

Worldwide Interpretability for Microwave Access (WiMAX) is a competitive connection oriented technology for metropolitan broadband wireless access with very high data rate, large service coverage and flexible quality of service (QoS). Due to the large number of connections, the efficient bandwidth management and related channel allocation for the uplink access in WiMAX networks is a very challenging task of the medium access control (MAC) protocol. In order to provide better bandwidth utilization and network throughput, a cost-effective WiMAX bandwidth management scheme is devised, named as the WiMAX partial sharing scheme (WPSS) and compared against a simpler scheme, named as the WiMAX complete sharing scheme (WCPS). An analytic maximum entropy (ME) model is proposed for the cost-effective performance evaluation of the two bandwidth management schemes associated with networks with a large number of stations and/or the connections. In this context, an open queueing network model (QNM) is devised, consisting of interacting queues with multiple service classes and closed form expressions for the state and blocking probability distributions are derived. Typical numerical examples are employed to verify the credibility of the analytic ME solutions against simulation results at 95% confidence intervals. Moreover, it is shown that the WPSS can significantly improve the networks performance in comparison to that of the WCPS.