An improved algorithm based on network load prediction for 802.11 DCF

In order to improve the performance of IEEE 802.11 Distributed Coordination Function (DCF) on throughput and packet loss, a novel backoff algorithm based on network load prediction is presented in this paper. The new algorithm firstly introduced a parameter to describe the network load condition. Then the prediction processing is implemented using the parameter sequence to get a predictive value. At last, a slot selection program is proposed based on the predictive value to overcome the bad influence caused by the uneven slot selection. The new algorithm reduces the collision probability of sending data packet by nodes, which does not need to estimate the number of network nodes, and can adaptively make the slot selection to send data packet by the node itself according to the network load level. Theoretical analysis and experimental results show that the new algorithm can optimize and adjust the network conditions automatically in different scales of network, and elevate the network performance on throughput, packet loss rate, energy efficiency and other property.