Research On MAC Layer Backoff Algorithms In Dense Wireless LANs

The performance of WLAN depends on the performance of the physical layer and the MAC layer.The most common way to improve the performance of the MAC layer is to modify the backoff algorithm,in which the commonly used method is to modify the backoff rule and adjust the competition window.The BEB algorithm in IEEE 802.11 reduces the collision probability when the station sends data frames according to the size of the Contention Window(CW),but when the number of participating stations is large,the CW value will fluctuate back and forth,resulting in reduced network throughput and increased delay.After analysis,it can be found that when the data is sent successfully,the competition window value is the window that has not collided after being adjusted,so this is a more suitable CW value.Therefore,if the current CW value stays after successful transmission,the collision probability can be reduced,which improves throughput.However,when there are fewer network stations,a large competition window will cause unnecessary waiting time.At this time,a small competition window value will reduce the waiting time of the station,thereby improving network efficiency.Taking the above factors into consideration,in order to improve the WLAN performance in a dense stations situation,this dissertation proposes a Probability Based Contention Window Retention Backoff(PCWRB)algorithm.In the PCWRB algorithm,when a station sends data successfully,the CW value of the station is set according to the probability: either the competition window value is set to the minimum value with a certain probability,or the current competition window value is set with a certain probability size,which avoids the throughput reduction caused by the CW value adjustment.The calculation and simulation results show that under both access modes,the performance of the PCWRB algorithm in throughput and delay is basically better than that of the comparison algorithm,and it can achieve relatively satisfactory results under dense network high load.Therefore,the PCWRB algorithm is a backoff algorithm that can be applied to dense WLANs under high load conditions.In the PCWRB algorithm,when a station transmits data,the backoff process of another station in the wireless coverage may still be in progress,and the data is also started to be sent at the end of the backoff,since the backoff time is often much shorter than the time of data transmission.When the data of the previous station is not sent,and the latter station has started to send data,causing collision between them,this will greatly reduce the throughput of the network.Therefore,based on the PCWRB algorithm,this dissertation proposes a Probability based Contention Window Retention Backoff with Freezing(PCWRBF)algorithm.In the PCWRBF algorithm,if participating stations detect that the channel is busy during the backoff process,then freeze the backoff time,suspend the backoff process,and monitor the channel until the channel becomes free again,the station activates the frozen backoff process,and the previous backoff count value is reused to start backoff.The calculation and simulation results show that under both access modes,the performance of the PCWRBF algorithm in throughput and delay is better than the comparison algorithm.Therefore,the PCWRBF algorithm is a backoff algorithm that can be applied to dense WLANs under high load conditions.