Channel Assignment And Node Localization Technology In Wireless Mesh Network

Compared with the traditional networks,wireless mesh network can be achieved dynamically self-configuring,self organization and self healing,the distributed network has higher capacity and data rate,the corresponding also easier to network maintenance,the network architecture can be more stable for commercial coverage or other more complex terrain environment,is the different network access method of fusion in the past,can think of wireless mesh network will be the next generation wireless network architecture foundation.However,with more and more in-depth research on wireless mesh network,many problems have been found in it,such as higher end-to-end delay after multiple hops,lower throughput caused by expanding communication range,and difficulty in achieving expected QoS(Quality of Service,QoS)of network service quality.Therefore,research on the key technologies of wireless mesh network to further improve its performance,give full play to the characteristics of wireless mesh network,and provide more secure network foundation for the upper layer applications has become the focus of current research on wireless mesh network.In this thesis,the channel assignment technology and node location technology in wireless mesh network are studied in depth,and are analyzed and compared with the existing algorithms based on the hybrid Wireless mesh Network.The channel assignment technology is studied to improve the overall capacity of wireless Mesh network,reduce channel conflicts and interference between links,and improve the resource utilization efficiency of network more effectively.By studying the node location technology,the range-independent location algorithm without additional equipment and extra cost for large-scale wireless Mesh network is realized,and the network connectivity information is used to locate the nodes.In this thesis,a link weight-based minimized collision channel assignment algorithm is proposed.According to the structure characteristics and traffic types of the actual hybrid wireless Mesh network,the definition and calculation method of link weight,collision coefficient and link collision are proposed.Algorithm considering the existence of network flow of different types,and try our best to save the network topology structure,through all the links in the network weights are calculated respectively,according to the weight to calculate the network link in the conflict,and thus to minimize network conflict channel allocation,can effectively reduce because of the network capacity attenuation caused by interference and channel conflict,and gives the complete algorithm flow and channel allocation mechanism.The overall network performance simulation was carried out with the network simulation software ns-3,and the improvement of the overall network throughput performance of the proposed algorithm and the existing algorithm under different business scenarios was compared and analyzed.In this thesis,a range-free node localization technology based on machine learning regression model is proposed.Based on the most classical range-independent dv-hop localization algorithm and its various improved algorithms in recent years,the modified hop-vector node localization method(HOP-POS localization method)based on machine learning regression model is proposed.The proposed algorithm directly uses the hops information of unknown nodes arriving at each anchor node in the network as the features of unknown nodes,and takes the actual location of nodes as labels to conduct the regression relationship modeling of node hops vector features and node positions.The performance differences of HOP-POS localization methods proposed by different machine learning regression models are compared.The proposed HOP-POS node location method can effectively combine with the existing machine learning regression model to greatly improve the node location speed,and at the same time in the network node density is relatively high,the number of anchor nodes is large can e ffectively improve the accuracy of node location.Finally,the research work in this thesis is summarized,and the future optimization of the proposed channel assignment and node localization algorithm is prospected.