An Efficient Scheduling for Ensuring QoS in WiMAX

Abstract

Broadband Wireless Access (BWA) has been serving enterprises and operators for years, to the great satisfaction of its users. Enterprises, Internet Service Providers (ISP), and Mobile Network Operators are all looking for cost-effective ways to move voice and data amongst multiple, separate locations at broadband speeds. WiMAX is an emerging broadband wireless technology and it has the potential to be a viable alternative to traditional wired broadband techniques as well wireless broadband techniques mainly due to its cost advantage. WiMAX supports multimedia applications like voice over IP (VoIP), Video Streaming, Online Gaming etc. These applications are diverse in nature and they have diverse requirements to be satisfied. It is necessary to provide Quality of Service (QoS) guaranteed with different characteristics for BWA networks. Therefore, an effective scheduling is critical for the WiMAX system. Many traffic scheduling algorithms are available for wireless networks, e.g. FCFS, Round Robin, Proportional Fairness (PF) scheme and Cross-layer Scheduling. Among these conventional schemes, some cannot differentiate services, while some uses different algorithms for different classes.

This thesis makes a comparative study of various available scheduling algorithms based on QoS criteria. Cross-Layer Scheduling approach is better as it is priority based and priority is calculated dynamically based on various service specific parameters. We propose a modification to Cross-Layer Scheduling algorithms so that efficient utilization of resources is also possible along with QoS. By extensive simulation, it is shown that the proposed modification delivers better results compared to original algorithm as far as efficient utilization of resources is concerned while ensuring QoS. The proposed algorithm also performs fair allocation of resources among different connections so that low priority connections get better service without affecting QoS requirements of other higher priority connections. Simulation results are also extended over period of time as well as number of connections to demonstrate scalability of proposed modification.