Using Hopfield Neural Network to Improve the Performance of Multi-rate WLANs

摘要

In a multi-rate WLAN environment, the trade-off between fairness and throughput is a problem, in this paper, we will construct a new intuitive simplified mathematical model called simplified coefficient of variation (SCV) model, through controlling the power of Access Points to optimize and improve the balance between proportional fairness among users and network throughput in multi-rate 802.11 WLANs. Since this trade-off problem is NP-hard, we use hopfield neural network solution to solve our model in a practical scenario. The simulation gives excellent results that indicate our model is effective and superior to existing methods. After the experiment analysis, we use software SAS to reveal the relationships among the various parameters.