Research On Wireless Network Coverage Optimization Algorithm Based On Big Data Analysis And Visualization Platform Design

With the rapid development of wireless network,more and more mobile devices are gradually accessing the wireless network,which makes mobile services in the communication network explode.However,some areas in the city are gradually being exposed to problems in network coverage.Better signal coverage is a prerequisite for ensuring mobile communication services.Therefore,ensuring good coverage of base stations is one of the important tasks of communication operators.Aiming at areas with coverage problems in the LTE network,traditional network optimization work mainly relies on engineers to adjust parameters on the site to improve regional coverage.This method is inefficient and poor in accuracy.This thesis combines actual LTE network drive test data with big data analysis methods,and deeply studies the LTE network overlap and weak coverage optimization methods based on big data.Based on the introduction of LTE network characteristics and existing network coverage optimization methods,this thesis mainly focuses on the following three aspects:(1)A method of LTE network overlap coverage optimization based on big data mining is proposed.The traditional overlap coverage optimization modeling method assumes ideal conditions,and there is a certain gap with the real LTE network environment.The method proposed in this thesis is based on the actual drive test data of the LTE network.First,the measured data is cleaned and expanded,and then the random forest algorithm is used to extract the important base station parameters that affect the regional overlap coverage.After the parameters are adjusted,the support vector machine algorithm is used to predict the adjusted area overlap coverage.Simulation results show that this method reduces the overlap coverage of the optimized area from 45% to 27%,effectively reducing the overlap coverage of the area.(2)A method for optimizing LTE network weak coverage based on the corrected propagation model is proposed.In the research of network deployment and network optimization,the propagation model can be used for coverage prediction.Aiming at the problem that the traditional propagation model is no longer applicable to the actual communication environment,this thesis first uses actual drive test data to correct the SPM propagation model,and then finds the optimal base station parameters based on genetic algorithms.In the process of finding the optimal base station parameters,the corrected propagation model is used to predict the coverage.The simulation results show that this method reduces the weak coverage rate of the optimized area from 17% to 0%,achieving full coverage of the area.(3)Designed a visualization web platform for the network planning system and realized network data visualization.The platform is based on the theory of data visualization,builts with the Vue framework,and uses Echarts components to draw actual drive test data into rich charts.The platform uses actual data in Jiangning area of Nanjing as a case to show the distribution of base stations,coverage of base stations,weekly downlink bytes of base stations and weekly average throughput of base stations in Jiangning area of Nanjing,it reflects the operation of base stations in Jiangning,Nanjing from multiple dimensions.It is helpful to the visualization of network planning and optimization work of communication operators.