The Research On Distributed Storage And Retrieval Technology Based On Massive Radio Monitoring Data

With the wide use of Internet of things,4G network,mobile Internet and other technologies,radio communication is widely used.However,due to the large number of illegal base station,Illegal radio,Satellite TV interference device and other equipment,the order of the radio waves is severely disrupted.In order to guarantee the order of radio waves,it needs to monitor radio signals in time,monitor radio equipment used to disrupt order and punish the relevant personnel severely.Radio resource management is related to national security and privacy protection etc.With the continuous development of social information,the management and utilization of radio frequency spectrum resources are becoming more and more important.Radio monitoring network has formed a national,provincial and municipal monitoring center and has the ability to monitor network.At the same time,radio monitoring has formed a huge amount of monitoring data.However,because the radio station is dispersed and the network linkage ability and intelligence of the monitoring software are not high,the collected data are stored in the local database in the form of the original data without analyzing and sorting effectively.In addition,the SQL query efficiency is very low faced with massive monitoring data,which will lead to isolated storage of radio monitoring data,very low utilization rate of radio monitoring data and can't quickly provide the effective information to the radio regulatory department.In view of the above situation,this thesis presents a model of processing,storage and retrieval of distributed and massive radio monitoring data,which implements distributed storage and efficient retrieval of radio monitoring data.The main works of this thesis are:(1)This thesis presents a preprocessing method for radio monitoring data.Firstly,radio monitoring data monitored by radio monitoring equipment is divided according to the time dimension.Then the K-Means algorithm is used to cluster the monitoring data in the time dimension.Finally,the distance between the data and the class center point is calculated,and the noise points will be removed.(2)This thesis proposes a scheme to store and retrieve the characteristic parameters.This thesis extracted 33 characteristic parameters of the monitoring data in the time dimension according to some statistical characteristics of the radio signal and the connection between the signal's service types,which is used as the representation of all monitoring data during this period of time.(3)This thesis designs and tests a distributed storage and retrieval prototype system for massive radio monitoring data.It includes the design of UIP unified interface platform,storage layer,retrieval layer and time synchronization service,etc,uses NoSQL database HBase to realize the distributed storage and retrieval of the monitoring data and gives up the SQL database,which is now widely used by the radio regulatory industry.This thesis completes the design of distributed processing,storage and retrieval system for radio monitoring data and realizes the fast processing of radio monitoring data,which provides an orderly storage,fast search and reliable analysis of the monitoring data for the radio regulatory department and can greatly meet the future needs of mass monitoring data storage and retrieval.Besides,some research results of this paper have been successfully applied in the project of "radio intelligent monitoring network system" and "radio intelligent monitoring network system test network integration project based on provincial and municipal levels ".