Efficient back-off mechanism for multimedia support in IEEE 802.11E

The IEEE 802.11e standard of Wireless Local Area Network (WLAN) has been designed for improving the Quality of Service (QoS) of real-time applications. The back-off mechanism used in MAC layer of this standard cannot be adjusted dynamically in the event of network situation change. This research attempts to look into ways to produce an effective back-off mechanism that is adaptive dynamically to the network status and able to support QoS for realtime applications over wireless ad-hoc networks based on the IEEE 802.11e standard. The current research proposes a new algorithm so-called Dynamic Fast Adaptation of back-off algorithm for contention-based EDCA (DFA-EDCA) mechanism. The main concept of the DFA-EDCA algorithm is to use exponential functions to tune the back-off parameters adaptively according to changes in network load and serve the time-bounded multimedia applications rapidly. In addition, the DFA-EDCA algorithm also provides an intra-AC differentiation mechanism to increase the back-off time randomness and achieve discrimination of the same traffic priority on different stations. The proposed algorithm has significantly reduced both collision rate and packet delay simultaneously with an obvious increment in both system goodput and channel utilization ratio which leads to the quality improvement of multimedia applications. The performance evaluations are conducted by using NS-2 simulator. The simulation results demonstrate that the proposed algorithm has greatly outperformed the previous mechanisms such as the non-linear dynamic adaptation scheme of the minimum contention window (CWmin HA), dynamic adaptation algorithm of the maximum contention window (CWmax adaptation), Adaptive Enhanced Distributed Coordination Function (AEDCF), Random adaptive MAC parameters scheme RAMPS and the conventional EDCA mechanism. The results show that proposed DFA-EDCA scheme has significantly decreased the collision rate in the whole network by 34.6 %, and reduced the mean audio and video delay by 18.5 % and 20.8 % respectively compared with CWmin HA scheme in the heavy load network. It also improves the goodput of the system by 19 % and the channel utilization ratio by 10.6 %. On the other hand, in the light load network, the DFA-EDCA improves the total throughput by 7.1 % and the total end-to-end delay by 8.3 % compared to RAMPS scheme