Whole House WiFi Network Quality Analysis Algorithm And Application

With the rapid development of the Internet,WiFi technology has become the key to solve the indoor network coverage,and whole house WiFi network coverage has gradually become an essential part of every home,so operators need to maintain a good whole house WiFi network coverage for users at all times,but if operators regularly conduct whole house WiFi network quality inspection for all users,it is not only timeconsuming However,it is not only time consuming,but also requires a lot of financial support.Therefore,it is important to design a comprehensive and effective whole-home WiFi network quality analysis system.To solve the problems in the existing work,this paper conducts algorithm research and application development on the problems of detecting the number of mosaics of video stream transmission frames under whole-house WiFi and the feature representation and classification of each terminal indicator detection data,proposes a comprehensive evaluation method and intelligent analysis model of whole-house WiFi quality based on network video analysis and terminal indicator detection data,and designs and develops a whole-house WiFi network quality analysis system is designed and developed in the face of the actual needs of operators,and the effectiveness of the system is verified on the data set provided by operators.The work in this paper specifically includes:(1)A whole-house WiFi quality evaluation method based on network video analysis.In this paper,we study the ratio of mosaic generation during the transmission of detection video under whole-house WiFi,dig into the mosaic characteristics of video stream transmission frames,and propose a whole-house WiFi quality evaluation method based on network video analysis.In the process of user transmission of detection video,some of the video frames are obtained,and the number and proportion of mosaic sizes are calculated by scanning and detecting the video frames using templates of various sizes by means of pattern recognition to assist in confirming the network quality.(2)Comprehensive evaluation method of whole-house WiFi network quality.In this paper,we study various terminal index detection data under whole-house WiFi,mine the terminal data characteristics,and propose a comprehensive evaluation method for whole-house WiFi network quality.The operator’s big data platform is used to obtain the batch terminal index detection data,and the batch data is cleaned and pre-processed,and the hierarchical analysis method(AHP)and the distance superiority solution(TOPSIS)based on the entropy weight method are used to model the data in order to obtain the comprehensive rating of whole-house WiFi network quality and solve the problem of comprehensive evaluation of whole-house WiFi network quality.In this paper,the comparison experiments on the batch data evaluation under the batch data provided by the operator show that the comprehensive evaluation method modeled in this paper has significantly improved the results compared with the original comprehensive evaluation method of the operator.(3)Whole-house WiFi network quality intelligent analysis method.To address the rapid processing of subsequent batch data,this paper proposes a whole-house WiFi network quality intelligent analysis method.The method adopts a metric-based meta-learning method to achieve small-sample classification of terminal superiority and inferiority in response to the difficulties of operator data acquisition,the small amount of acquired batch data,and the characteristics of acquired terminal index data.In this paper,the broadband with high confidence in the whole-house WiFi network quality comprehensive evaluation method is labeled,and the batch with labeled data is pre-processed to carry out a pre-training process for the network to improve the differentiability of terminal superiority and inferiority categories and establish clear decision boundaries for the subsequent metric feature space distance.In the testing phase,only a small number of superior and inferior terminal data samples are needed to batch the terminal superior and inferior categories,and finally the whole-house WiFi quality is judged based on the number of superior and inferior terminals under whole-house WiFi.The experimental results show that the method has certain feasibility.(4)Whole-house WiFi network quality analysis system.Based on the requirements of operators,this paper develops and deploys a whole-house WiFi network quality analysis system,which integrates a comprehensive evaluation method for whole-house WiFi network quality,and performs offline batch calculations through Hadoop,a big data framework,and Hive,a data warehouse.The experimental results show the accuracy of its calculation under the data provided by the operator,while the availability of the system is ensured under the stress test on the system.