Research On 5G-SON And BeiDou Integrated Positioning Technology

Due to its inherent vulnerability,global navigation satellite system(GNSS)is difficult to provide available PVT(positioning,velocity and timing)services in indoor,underground,urban and other scenes.With the continuous development of the fifth generation mobile communication technology(5G),communication and positioning technology based on 5G is becoming more and more important in people’s production and life.5G positioning technology makes up for the inherent defects and deficiencies of GNSS to some extent.Satellite navigation and 5G mobile communication are important components of the national comprehensive PNT(Position,Navigation and Time)system,the integrated positioning technology of them has become a hot issue in current scientific research and engineering application.5G mobile communication allows direct connection and interworking between user terminals in the form of self-organizing network(SON),that is,the so-called terminal D2D(Device to Device)technology.Different from the mode of using 5G base station to locate user terminals,D2 D positioning uses the self-organizing network composed of mobile communication terminals to realize the absolute positioning between network nodes.The navigation and positioning technology between 5G-SON network nodes combined with satellite navigation will have a great impact and change on people’s production and life in the future.Taking 5G-SON and Bei Dou integrated positioning technology as the research object,this thesis has carried out and completed the following research:(1)The positioning principle of Bei Dou satellite navigation system(BDS),single point positioning technology using pseudo range,doppler velocity measurement technology and RTK(real time kinematic)high-precision positioning technology are systematically studied;This thesis focuses on the kalman filter algorithms commonly used in integrated navigation system,including static kalman filter model,dynamic kalman filter model,fading kalman filter,robust adaptive kalman filter and other filter algorithms,as well as the federated filter structure commonly used in multi-source information fusion.(2)The location algorithm between nodes in 5G-SON network based on ranging information is studied,and the solution equation of location information between network nodes is deduced;The various spatial coordinate systems used in the subject research are introduced in detail,and the transformation relationship between affine coordinate system used in SON system and Bei Dou coordinate system(BDCS)is studied;The verification and test platform of 5G-SON network node positioning algorithm is constructed.And the positioning performance of 5G-SON network nodes by using adaptive filtering algorithm in the static scene is tested and analyzed.The results show that there can achieve the decimeter positioning accuracy between network nodes in horizontal plane.(3)The framework of BDS / 5G-SON loose integration system is proposed,and its mathematical model is studied.The navigation and positioning performance of single BDS,single 5G-SON system and BDS / 5G-SON loose integration is demonstrated theoretically.This thesis analyses and compares the positioning performance of different positioning algorithms and different positioning systems in stationary and moving states in detail through experimental verification.The test results based on the measured data show that the performance of robust fading adaptive filter algorithm applied to a single navigation system is better than the traditional weighted least square method and kalman filter algorithm.Compared with the BDS single system,the root mean square error(RMSE)values of the BDS / 5G-SON loose integration system in the static scene are increased by 56.36%,0.80% and 6.51% respectively in the east,north and up directions,and in the dynamic scene are increased by 31.63%,16.62% and 54.02% respectively in three directions,the loose integration system can improve the accuracy,robustness and reliability of the BDS single system.(4)The framework of BDS / 5G-SON tight integration system is proposed,and the mathematical model is deduced.The positioning performance of BDS / 5G-SON tight integration system in static and dynamic scenes is tested,while comparing the performance of BDS / 5G-SON loose and tight integration systems under the same observation conditions,and a tight integration positioning model and algorithm assisted by long short term memory network(LSTM)are proposed.The test results based on the measured data show that: compared with the BDS single system,the error RMSE values of BDS / 5G-SON tight integration in the static scene are increased by 82.69%and 38.35% in the east and north directions respectively,and in the dynamic scene are increased by 56.42%,30.28% and 90.74% in the east,north and up directions respectively.The overall accuracy and robustness are better than that of single BDS system.Under the same observation conditions,two kinds of combination have similar positioning accuracy and can be applied to different scenes.Moreover,when BDS has few visible satellites,such as only 2 to 3 satellites,BDS / 5G-SON tight integration system can still realize navigation and positioning function,indicating that the sustainability of tight integration system is higher.When the BDS satellite loses its lock,the auxiliary system based on LSTM can deal with the problems such as the interruption of integrated positioning and realize the continuous navigation in the weak observation environment.The maximum RMSE values of the prediction results are 1.25 m and0.84 m respectively when there are less and more training data,which shows the effectiveness and feasibility of the neural network auxiliary system.