Mobile Internet Traffic Analysing Of Cell Based On Cloud Computing

With the rapid development of Internet, the scale of the Internet continues to expand, and the Internet has already penetrated into every aspect of people’s lives. Now, with the widespread popularity of mobile terminals, mobile Internet has become an indispensable part of people’s lives. Using a variety of mobile Internet business and communicating with the others, subscribers generate a wide variety of network behaviors. If we say that a few years ago the Internet brought the colorful varieties of information, then in recent years, the rapid development of mobile Internet will make it more convenient and easier to access to the Internet, therefore subscribers can be free to access Internet at any time and any place more quickly and easily to get all kinds of information.At first, this thesis analyzes the importance of analyzing mobile internet subscribers’behavior, and describes the logical structure of the mobile Internet and the features of mobile Internet traffic analysis and so on. Then, it introduces the concept of subscribers’behavior analysis, etc., and elaborates the basic knowledge of mass data processing, difficulties and handling techniques, and data mining purposes, procedures and methods. Then, it details the plot flows statistical methods, and does detailed comparative analysis of the relative errors of each statistical methods. The thesis also calculates the time complexity of each method, and does statistics and comparative analysis of the actual running time of each method. Because of the data set of Gn port exists error about cell reselection which affects the result of cell traffic statistics, this thesis firstly has the the basic features of a detailed statistical and comparison between the data sets of the Gb port and the Gn port. To determine the cell reselection on the data set of the Gn port traffic statistics obtained by the impact of the cell. Then, analys the cell traffic in multiple dimensions. And baed on data mining knowledge, having the cluster analiysis of cells of traffic using k-means algorithm, and the classification of the characteristics of each group showed an in-depth analysis and comparison.