Research On Positioning Technology Based On Correlation Sequencing In Distributed Antenna System

In recent years,with the popularity of mobile devices and wireless communication technology,location based service(LBS)using by teminal has attracted more and more attention,which has been widely used in many aspects including people's daily work and life,travel and business development.In outdoor environment,the global positioning system(GPS)can provide high-precision positioning services in real time,however,due to the loss of satellite signals caused by obstuctions from buildings,the positioning accuracy of GPS is declined seriously.At the same time,Distributed Antenna Systems(DAS),which has been used to satisfy indoor mobile communication,has achieved large-scale coverage.Hence DAS based indoor localization technology has become a research hotspot in recent years.In the DAS environment,the Reference Signal Receiving Power(RSRP)value received has the same antenna array element identifier in a certain area.Therefore,for the single dimensional signal strength data,the traditional fingerprint algorithm cannot be used to obtain the user's location.Based on aforementioned,this thesis designs a correlation sequence based indoor positioning algorithm in DAS system,which overcome the difficulty of DAS localization using single-dimensional fingerprints.Meanwhile,micro-electro-mechanical systems(MEMS)sensors are used to assist the construction of dynamic fingerprint database,which solves the problems of the locating difficult by using single-dimensional fingerprints and the high cost of fingerprint database construction.Therefore,it is mainly studied from the following aspects:1)The algorithm of skeleton extraction is studied to find the core path of the target scene,and then reduce the overhead of the off-line fingerprint database.Firstly,the topological graph of location scenes is drawn.Secondly,the topological graph is processed by binary image and the skeleton is extracted by using the thinning algorithm for the binary image processed.Thirdly,the skeleton serves as the core path of the target scene which is used to collect data and build the moving sequence fingerprint database.2)Studying the correlation sequencing algorithm is used to solve the problem that single dimensional signal is difficult for localization under DAS.First,the matching degree matrix is calculated to determine the set of alternative location sequences.Then using backtracking method to find matching segments and calculate the correlation coefficient and average Euclidean distance.Finally,the matching degree,correlation coefficient and average Euclidean distance are adaptively chose using the voting method and the optimal matching sequence can be obtained for realizing the position estimation of the target.3)On the basis of the algorithm design,tthe DAS positioning system is built,and the feasibility and effectiveness of the algorithm are verified in a variety of scenes.The system uses the existing public network signals and hardware devices to achieve cross floor localization,and the average positioning accuracy is about 3 meters.The proposed system has a good practical value.