Research On Positioning And WIFI Coverage Optimization Based On Network Traffic

In the era of mobile Internet,the quality of network services is closely related to people’s lives.Traditional methods use physical layer information to study WIFI indoor positioning and WIFI coverage identification,but these information are often encapsulated and encrypted and cannot be directly obtained.It has become a fast and effective method to study WIFI indoor positioning and WIFI coverage between operators and users through the traffic data provided by the mobile network.The main contributions of this paper are as follows:(1)WIFI indoor positioning algorithm.A KNN algorithm based on coefficient correction is proposed to solve the problem of low positioning accuracy of the traditional K-Nearest Neighbor(KNN).First,the traditional KNN algorithm uses the Euclidean distance to measure the distance without considering the data correlation.By calculating the Spearman correlation coefficient between the location fingerprints,the data that do not meet the correlation are screened.And then the traditional weighted K-Nearest Neighbor algorithm(Weighted K-Nearest Neighbor,WKNN)uses the inverse of the Euclidean distance as the weight of the distance metric without considering the data stability.By changing the weight to the inverse of the variance,data with higher stability get higher weight.The experimental results show that,compared with the KNN algorithm and the WKNN algorithm,the algorithm proposed in this paper improves the traffic location accuracy by 32% and 11%,respectively.(2)Identification method of WIFI coverage.In view of the problem that traditional methods cannot identify dynamically,a method for identification of WIFI coverage based on network traffic is proposed.First,capture a large number of network traffic packets and perform preprocessing and feature selection.Then,the obtained features are input into a variety of different machine learning models for training,and the model with the highest recognition accuracy is obtained.Finally,the data packets to be identified are input into the above model after the same processing to obtain the WIFI coverage.The experimental results show that among different machine learning models,the Transformer model obtains the best recognition accuracy,and can achieve an average accuracy of 88.8%.(3)WIFI coverage location optimization.The proposed method in(2)can only identify the WIFI coverage of the indoor environment and cannot optimize the coverage considering the user’s network requirements.This paper proposes an improved gray wolf optimization algorithm based on traffic weights(Improved Grey Wolf Optimization algorithm based on Flow Weight,IGWO-FW).The population initialization is more dispersed through chaotic optimization,and then the dynamic fusion mutation mechanism is used to avoid premature convergence of the algorithm to the local extreme value and improve the efficiency of the algorithm.The simulation results show that the IGWO-FW algorithm can effectively capture the traffic weight.Under the same number of routers,the proposed algorithm can improve the network throughput by 16.7% compared with the network traffic of traditional GWO algorithm.