Design And Implementation Of Performance Management And Analysis System For Mobile Stations

With the mature of LTE network technology, LTE equipment have being deployed to existing GSM and TD-SCDMA mobile network, increasing the difficulty of network performance optimization for mobile communication company. Since the three kinds of network coexist, it should take three kinds of network performance into account when the mobile network parameters is being optimized and adjusted or equipment is to be planning and collocation, but the network equipment manufacturers only provide a single performance data for mobile company network optimization staffs, which gives them inconvenience to analysis and process for the whole mobile network performance.This paper has designed a mobile base station performance analysis system for centralized management of three kinds of networks for a mobile communication company, comprehensive analysis on three kinds of network devices' operation status with the system, helping workers to optimize the allocation of resources as well as the planning and design of mobile network, so as to enhance the quality of mobile network service. Mobile base station performance management analysis system is mainly composed of data import module, co-base station verification and statistics module, the cell performance analysis module, GIS sites positioning module and the presentation of data processing results, using both MFC client and WEB client for achieving the system tasks. Firstly, the system uses the mobile resource management data in mobile communication company for achieving the verification and statistics of all equipment at the same site, and it facilitates network optimization personnel co-base station equipment's performance analysis and inter-network's optimization. Secondly, the system achieves importing three network performance data into system rapidly, and multi-dimensional queries with graphical display for cell performance helps network optimization staffs intuitively understand the operation of network performance. Thirdly, to analyze the network congestion and excavate the high load cells, the paper uses robust minimum distance algorithm (LDM) to extract its linear trend based on cell wireless utilization ratio, then use the K-means cluster algorithm to classify all cells to help staffs grasp the overall network congestion. Fourthly, the paper uses the linear combination model of ARIMA and Elman for predicting the trend of important cells, and the prediction accuracy approaches a good fitting. It can effectively help staffs grasp the future trend of the important cells so as to adjust cell parameters timely for preventing overload of cell performance. Finally, the system achieves GIS sites module with the help of Baidu Maps and Ajax technology, and it helps staffs find the location of a base station quickly to facilitate regional performance optimization.The performance management and analysis system for mobile station has been on-site running in a mobile communication company, and the system has been steady running for six months by means of on-site debugging and function adjustment. The 900 000 daily cell performance data in allover network could be imported into system in 8 minutes, and the monthly assessment of the LTE cell congestion takes about 5 hours. System test results show that each module is running normally, and the user experience is good. While the linear combination model of ARIMA and Elman has a good prediction effect on cell performance trends, and it can help staffs master the dynamic changes of the cells.