Power Allocation For Multi-radio Multi-channel Network

Wireless Mesh Network(WMN)is a Mesh network based on wireless communication.Compared with traditional wireless local area network(WLAN),WMN has better performance in scalability,robustness,and throughput,etc.Due to the current increase in the number of wireless terminals accessing the network,how to effectively deploy the WMN network has become a research hotspot.The channel and power are two important resources,and how to use them effectively is a difficult problem that must be solved when deploying WMN.To solve the problem of low channel utilization in ordinary WMN,multi-radio multi-channel wireless mesh networks(MRMC-WMN)have become a solution.Current MRMR-WMN use static resource allocation such as average power allocation,which also results in its lack of good scalability.Therefore,how to flexibly adjust the network topology through power allocation to achieve high performance is critical.This thesis studies the power allocation problem under the condition of static channel allocation,and proposes distributed power allocation algorithms to improve throughput and fairness.The main research results are as follows:This thesis introduces the research background and significance of MRMC-WMN and makes a classification summary analysis of current power allocation algorithms.The above work provides theoretical support for the rest research of this article.To maximize the throughput of the entire network,this thesis first proposes a throughput model based on the random transceiver network model and designs a distributed power allocation algorithm for maximizing throughput.The algorithm is based on greedy rule and convex analysis and gives the theoretical optimal solution of power allocation.Through the performance comparison with other existing power allocation schemes,the interference,throughput,and other network factors of this algorithm are analyzed.According to the simulation results,the network throughput performance of this algorithm is higher than other existing algorithms,and it has good adaptability and convergence for different network scales.To maximize fairness and meet the QoS requirements of all links,a corresponding distributed fairness maximization power allocation algorithm is proposed in combination with Max-SR and Max-Min rule.By analyzing the non-convex mathematical model established by the two major rules,a non-cooperative game model with supermodel game properties is established.The above non-convex problem is divided into multiple sub-convex problems.The theoretical optimal solution is given by the sub-gradient method.The author builds a simulation platform based on MATLAB,and gives the simulation results under different networks.The results show that compared with other algorithms,this algorithm can effectively improve the fairness of the network.