Evaluating saturation throughput performance of the IEEE 802.11 MAC under fading channels

mhempel2's photo

Publication Type:

Conference Paper

Source:

Broadband Networks, 2005. BroadNets 2005. 2nd International Conference on (2005)

Keywords:

access protocols, admission control, carrier sense multiple access, carrier sense multiple access-collision avoidance, CSMA-CA protocol, DCF, fading channel error, fading channels, IEEE 802.11 MAC, Markov chain, Markov processes, medium access control, Ns-2 simulation, QoS provisioning, quality of service, resource allocation, scheduling, scheduling scheme, telecommunication congestion control, wireless LAN

Abstract:

Accurately modeling the IEEE 802.11 MAC is critical to design resource allocation, admission control and scheduling schemes for QoS provisioning in wireless LANs. However, it is difficult to model the saturation throughput performance of the IEEE 802.11 MAC due to the fact that the memory effect existing among different stages in the random backoff procedure. Therefore, Markov chain has been used to characterize the saturation throughput performance of the IEEE802.11 CSMA/CA protocol in many research efforts. However, one of the problems existing in the current research failed to consider all key practical issues altogether such as retry limit, backoff suspensions and the fading channel errors. Therefore, an accurate and realistic model of the IEEE 802.11 MAC protocol with fading channel errors is needed. In this work, we propose a new model based on Markov chain and validate its correctness by theoretical analyses and Ns-2 simulations. We also compared our results with other current major models. Our contributions in this work are: 1) taking the backoff suspension into considerations and derive a more accurate model than other current models; 2) introducing the fading channel errors into the modeling process and derive a close form equation of the saturation throughput performance of the 802.11 DCE simulation results show that the proposed model is same as accurate as other current models, and becomes much more accurate when considering fading channel errors.