| Telecom value-added services have become the major income source of telecommunication operators. In the market driven by users' requirements, the view of customer-centric and treating maintenance of Network and services quality as core function is the key points in Network Operation Support System's development. So that Intelligent Network Operation Quality Analysis System is designed to meet such demands as service-oriented quality assurance.EB Quality Analysis System is designed and implemented aiming on integrated and smart management of Intelligent Network and Services. Its main functions includes collecting and analyzing NE's(Network Element) running state and services operation information or some data like that, building suitable KQI-KPI(Key Quality Indicate, Key Performance Indicator) indicator hierarchy model, so as to produce statistics and give a comprehensive evaluation of the system operation quality. EB-QUANSY is flexible and meet for the various maintaining demands of telecom operators.Fault correlation and trend analyzing function modules are relatively independent but also certain associated with each other in EB-QUANSY. Both in order to maximize the effective analyzing work, reduce the occurrence of bad failure impact, timely detecting and evade failures, also, indirectly improve the management of operation and maintenance work efficiency and reduce man-made causes. Especially, Fault Correlation is the premise of automation and intelligent Troubleshooting. Through a variety of types of data correlation analysis, extraction of effective and useful associated rules, and applied them to deal with the actual fault situation, can provide assistance for engineering with fault diagnosis and better troubleshooting and prevention capability. However, Trend Analysis function is based on the needs of accessing to know the possibility of future state; it must to set on the base of sophisticated analysis techniques and with use of math arithmetic or logical methods to find the regular pattern of historical data, so as to help people in control of the current tendency or process. Therefore, the two features combined acting to provide a simple decision-making and guidance for network expansion, business development, etc.In the process of fault correlation function and trend analysis function design and implementation, large amount of experiences are accumulated, and also gaining some features of the two and applicable formulas for extracting correlated rules and for trend analysis performing, and some guiding principles could also be deduced. With the function modules structure and design principles summarized, it will later being discussed individually as follows.This thesis focuses on the design of fault correlation and trend analysis functions in EB-QUANSY system, it is divided into several sections as follows: First of all, an outline of EB-QUANSY system is given, which is the support framework of all functions; then the relevant background knowledge is introduced, including Data Mining and associated rules mining technologies and time series prediction algorithms. In Chapter 2, key issues encountered during the development are analyzed, and methods or solution for each issue are described. Based on this, features of fault correlation and trend analysis functions are summarized. In Chapter 3, the design and implementation schema for the two function modules are proposed, with detailed description of function each task unit in module structure, the communication message flow and as well as the application of key technologies have been conducted. In Chapter 4, give some testing cases of the two functions in real test environment with results comparison of them added before and after. Giving discuss about the comparison, presenting charts and diagrams to illustrate system's performance optimization. The last chapter analyzed the future development of Telecommunication Operation and Maintenance System and the future application for typical fault correlation and trend analysis functions. |
PDF Full Text Request