Research On User Behavior Analysis Based On 4G Operation And Maintenance Data

4G technology,making mobile communications into a faster era.LTE communication system IP-based data transmission has changed the original communication methods,including advanced technologies such as OFDM,MIMO,smart antennas,making LTE systems more compatible,wider coverage,cheaper tariffs and a stronger received signal.Statistics released in May 2016 show that the country's 4G users have reached 530 million,accounting for 41% of the entire mobile communications market,Premier Li Keqiang in 2017 two government work report pointed out that in 2016 4G users net growth of 340 million.In less than four years,4G communication in the mobile communication has occupied a dominant position,due to the short time,4G data mining research is still in the incomplete stage,so the 4G technology and its derivative field Analysis is necessary.The thesis is based on these advantages,the analysis of the user's travel behavior and group aggregation behavior.In this thesis,we use the RSSI and centroid localization algorithm to locate the user's position,and propose an improved clustering algorithm to study the behavior of the user from the individual and the group.At first,we make a detailed analysis and design of the system requirements,the overall system architecture and the database,and introduce the structure and operation principle of the LTE communication system,analyze the main network elements of EPC and the key interface protocol of extracting the source data,we get the source of 4G operation and maintenance data in this thesis,,and then preprocess the results of the analysis to filter out the source data needed like the user's location,time,and business.Secondly,followed by the analysis of the contents of the operation and maintenance data again with targeted processing,excluding isolated base stations,we combine RSSI positioning and centroid positioning of the two algorithms,proposed R-centroid localization algorithm in this thesis,through the positioning algorithm using the positioning point of the sequential tracking the user's travel path.Then,the author analyzes the individual behavior of the individual user by analyzing the travel characteristics of the individual user,and puts forward the travel characteristic index,analyzes the number of trips and travel distance in the individual user's trip,and uses the clustering algorithm to extract the user's stay point to the individual user.In the user analysis,the clustering algorithm based on the stay point K-means and the hierarchical union is analyzed,and the similarity analysis of the user behavior is analyzed.The aggregation point of the population is analyzed,and the aggregation behavior and aggregation purpose of the user are analyzed more precisely.Combined with 4G to obtain the convenience of App business information for the algorithm to provide a reference,getting more accurate warning.Finally,the design and implementation of the system are described in detail.Combined with the test of the system function,the correctness of the result is analyzed,which proves that the system can be applied to the behavior analysis and has high accuracy.