A Fair Backoff Algorithm for 802.11 WLAN

• Main Authors: Athena Tang, 唐淑芬
• Other Authors: Chung-Ju Chang
• Format: Others
• Language: en_US
• Published: 2004
• Online Access: http://ndltd.ncl.edu.tw/handle/04987899178129533398

碩士 === 國立交通大學 === 電機資訊學院碩士在職專班 === 92 === When more and more people enjoy the convenience of the wireless LAN, we find the traffic loading will introduce the serious delay time of packets. Both the delay time and the standard deviation of delay time increase along with the traffic loading. This will induce the unfair packet transmission. According to the conventional backoff algorithm, a packet with more collisions will select a backoff timer from an exponential-expanded contention window based on the collision number, which means getting a larger backoff more possible if the packet is collided more. We wish to specify a fair backoff algorithm to balance all packet transmissions within a predictable delay time. The standard deviation of delay time should be as small as possible to achieve this goal. Instead of using the Uniform distribution, the Gamma distribution is the function we use in the fair backoff algorithm. The two parameters of Gamma function come from the measurement of the traffic loading and the collision number of that packet. From the simulation results, we can recognize that the standard deviation can be reduced 70% if comparing with the conventional backoff algorithm under any kind of traffic loading. We can enhance the precision of delay time without degrading the channel utilization or the average delay time.