Design And Implementation Of Maintenance System Based On BME Framework For Site Operation

In the process of information construction of communication enterprises,the maintenance for the site operation occupies an important position.As the number of services grows sharply and the complexity of services continues to rise,the degree of dependence of communication companies on the maintenance system is also increasing.A good maintenance system for site operation can minimize the operational risk of the communication enterprise and improve the operation efficiency.In recent years,along with the rapid development of internet technology,there are more and more maintenance systems developed for the business of telecommunications enterprises.However,there are still many problems in these systems.For example,the system requires technicians to master a variety of development technologies,which are time-consuming and labor-intensive;the systems of various departments operate independently,and the business data is not aggregated;the system does not simplify the process of business;the artificial detection of business data is too passive.Based on the practical experience and the above problems,this thesis designs and implements a maintenance system based on BME framework for site operation.This thesis is based on the BME framework for system development,which effectively improves the convenience of system development and maintenance.The system also aggregates the site data of various departments and supports one-click import,which solves the problems of data dispersion and tedious business operation processes.At the same time,this thesis designs and implements a solution for anomaly detection of site time-series data,which improves the efficiency of site business maintenance.The main work of this thesis includes the following aspects:(1)This thesis designs and implements a maintenance system based on BME framework for site operation.The system includes an information management module,an information visualization module,and an anomaly detection module.It improves the work efficiency.(2)The characteristics of the site time-series data are analyzed.The solution of the anomaly detection is designed according to the characteristics.In the solution,the existing 3-sigma-based dynamic threshold algorithm is improved,which effectively improves the accuracy of anomaly detection for time-series data with day periodic characteristics.(3)This thesis uses the Grafana tool to chart the site information,which enhances the visibility of the site information.