The Design And Implementation Of Cluster Analysis And Resource Optimi Zation System In Cellular Wireless Network

As the rapid development and wide application of LTE technology,mobile user behavior law has also changed greatly.The rapid development of wireless technology,bringing the mobile terminal uses frequency increases,and the using time becomes much long.Explosive growth of wireless data,bringing an increasing number of operational opportunities and data services,At the same time,the growth trend of mobile end data enables operators become more difficult to optimize network resources.Thus using new analytical methods to analysis of user,business and resources multi-dimensional data have great significance for wireless network planning and resource optimization.With the commencement of LTE data services,in order to cope with the massive surge in communication data types and quantities,our work use the data mining method in the cellular radio network data to processing and analysis.On the other hand,users of mobile devices to access the Internet through a growing frequency.Equipped with Android operating system device share of over 80%,Android OS has excellent representative.Based on the above background,the work analysis the cellular radio network users,services,resources data first,and gives an association prediction method based on multi-dimensional cluster analysis.Then building the mobile Internet terminal analysis system based on Android,realized the overall process of collection of data from the client,and use clustering algorithm analysis,and results show on the web side.The main contents are as follows:1.It gives a multi-dimensional cluster analysis based associated fitting prediction methods.In this method,the first is use principal component analysis to reduce the dimension of the sample data,followed by the process of the secondary cluster analysis of the data,and the results are fitted using a gradient,with the cellular radio network data associate forecast.Research indicates that compared with direct fit to the data,cluster analysis can be improved forecast accuracy on large number of LTE data,Optimize the aspect of resource forecasting.2.Designed the mobile Internet terminal systems based on the Android OS.The system includes data acquisition system based on Android terminal,and server-side data analysis system.Android terminal in the acquisition system can collect a variety of data cellular subscribers,including user basic information,signal data,call data,SMS data,traffic data and communications traffic data and other abnormalities data,and recording terminal operations occur when GPS coordinates at the same time.The server analysis system storage terminal acquisition data,implement the clustering algorithm in our article,and use the real network's data analysis and forecasting.This design analysis system,combined method presented in this work,offered a complete solution process in cellular wireless network for the data collection,analysis,forecasting.