الملخص الإنجليزي
IEEE 802.11 wireless local area networks (WLANs) are one of the most popular and most widely deployed WLAN technologies worldwide. The explosion in the use of wireless networks has lead to increasing demands for supporting the emerging multimedia applications, e.g. VoIP and streaming video. Such applications are growing rapidly and are often characterized as bandwidth intensive and delay sensitive. This has led to the requirement that 802.11 WLANs should provide Quality-of-Service (QoS) in order to efficiently support multimedia applications. As a consequence, the IEEE 802.11é standard has been released for QoS capabilities in WLANs by introducing a contention window-based Enhanced Distributed Channel Access (EDCA) technique. Although a number of research studies have shown that EDCA is able to provide QoS support, existing EDCA techniques reset the Contention Window (CW) of a given wireless station statically after each successful transmission. This static behavior does not adapt to network state, i.e., the congestion level inside the network, which often results in poor performance and low link utilization whenever the demand for link utilization increases. Several techniques that are based on adapting CW to network conditions have been suggested to enhance the performance of IEEE 802.1le. The first part of this thesis analyzes the differentiation capabilities of the important EDCA QoS parameters by means of throughput analysis in ad hoc WLANs. The main aim of such an analysis is to gain an insight into how the QoS parameters can be best applied to support QoS requirements. The second part of this thesis implements a number of enhancement schemes, notably Simple Static Slow Decrease (SSD), Collision Rate-based Adaptive EDCF (CR-EDCF) and Simple Recursive Enhanced DCF (SR-EDCF), that have been proposed for EDCF in OPNET, a popular simulation software for wireless networks, based on the original 802.11e model. The resulting simulation model is then used to conduct a performance comparison of the enhancements schemes, as well as the original standard scheme. Our results will show that SSD outperforms the other enhancement schemes and EDCF while maintaining relatively low complexity requirements. The last part of the research proposes an enhancement to improve the performance of EDCF by tuning EDCA QoS parameters. Our simulation results will demonstrates the effectiveness of the proposed enhanced SSD scheme (E-SSD) in providing high performance compared to SSD and the other enhancement schemes.
