Design And Implementation Of Business Support Systems For Communication Operators

The rapid development of next-generation communication technologies and the increasing popularity of mobile smart devices.Mobile communication networks are penetrating into all areas of people’s lives and have become an important basic information facility in today’s society.In the face of increasingly complex network environment and exponential growth of network data traffic,in order to effectively avoid network congestion and network blockage,communication operators’ business support systems need to spend a lot of time and resources to monitor real-time network traffic to cope with different unexpected situations.Based on the real traffic data from the base stations of HS city communication operators’ networks,the project investigates the theories and algorithmic ideas related to traffic prediction at home and abroad,and builds a Broad Forest model for traffic prediction management based on the communication operators’ service support systems.The model learns by introducing Random Forest(RF)in the framework of Broad Learning System(BLS),so as to predict the possible changes of network conditions in advance and take early countermeasures to ensure the stability and good service quality of the network.The main research work and contents of this thesis include the following.(1)System requirement analysis: Firstly,the feasibility analysis study of the system is conducted from three aspects: technical feasibility,economic feasibility and operational feasibility.Then the system is divided into five business subsystem modules: monitoring management,business management,analysis management,operation and maintenance management and resource management according to the specific business requirements of communication operators.(2)Model design: The data set used in the model is derived from the network traffic data collected from a real base station of an operator,but the original data cannot be directly input into the model,and operations such as removing outliers,data transformation,missing value filling,and feature extraction and feature selection need to be performed on the data.Finally,it is input to the Broad forest model for prediction,and the results are practical for an operator to predict the changes of base station traffic data.(3)System implementation: Adopt mainstream Vue+Spring Boot front and back-end separation development model,rely on B/S architecture,use Spring MVC and My Batis technology to build the system.Use ECharts library to visualize Oracle data,and deploy big data processing platform for data storage and integration to provide reliable data resources for model construction.(4)System testing: The implemented system is tested mainly from both functional and non-functional aspects to check whether the system functions effectively and whether it meets expectations in terms of security,compatibility and stability.The test results show that all aspects of the communication operator’s business support system meet the user’s expectations.After data processing of past data,the business support system of communication operators combines data storage technology and machine learning algorithms to make realtime prediction of different base station traffic data,and through continuous optimization of the model,eventually realizes a higher real-time prediction effect,allowing operators to predict network traffic-related changes in advance to ensure good service quality.By combining data storage technology and machine learning algorithms with data processing of past data,the business support system of communication operators makes realtime prediction of different base station traffic data,and finally achieves high real-time prediction effect by continuously optimizing the model,so that operators can predict network traffic-related changes in advance to ensure good service quality.