| TC-OFDM(Time&Code Division-Orthogonal Frequency Division Multiplexing)is a high-precision indoor and outdoor seamless positioning system.It achieves high-precision indoor and outdoor positioning services on the basis of non-interference of communication signals by means of communication navigation signal fusion,which is suitable for large-scale deployment.However,the indoor environment is complex and changeable.The location signal is affected by multi-path and non-line-of-sight in the transmission process,which results in low positioning accuracy and poor robustness.Inertial Navigation System(INS)is a positioning method which is less affected by the environment,but it can only achieve relative positioning,and is prone to cumulative errors.Therefore,TC-OFDM and INS integrated navigation can achieve complementary advantages and improve positioning accuracy and robustness of integrated navigation system.Traditional TC-OFDM/INS integrated navigation usually adopts loose coupling mode,which can not make full use of the measurement information of TC-OFDM receiver,nor can it assist the loop of TC-OFDM receiver.In this paper,a TC-OFDM/INS deep integrated navigation system based on multi-state constrained Kalman filter is proposed.Carrier phase observation is introduced and INS information is used to assist TC-OFDM tracking loop,which effectively improves positioning accuracy and stability.The main contents of this paper are as follows:(1)A TC-OFDM/INS integrated navigation system model based on multi-state constrained Kalman filter is designed.The integrated navigation model based on extended Kalman filter has Markov property,and its error is greatly affected by the previous positioning results.The complexity of indoor environment easily leads to the fluctuation of positioning results.To this end,a Kalman filter model with multi-state constraints is proposed.The model uses the measurement information of multiple time series to correct the state variables jointly.The time difference carrier phase is used as the observation quantity,and it can adjust the time synchronization error by feedback,so as to improve the positioning accuracy and stability of the system.The simulation results show that TC-OFDM/INS integrated navigation model based on multi-state constrained Kalman filter has higher positioning accuracy than that based on extended Kalman filter.(2)An INS-assisted tracking loop correction algorithm for TC-OFDM receivers is designed.Aiming at the problem that the dynamic and stability of TC-OFDM receiver tracking loop can not be taken into account and is seriously affected by indoor multipath,this paper designs an INS-assisted TC-OFDM receiver loop correction algorithm from two aspects of carrier tracking loop and code tracking loop.The carrier loop frequency and code loop phase discriminator are assisted by the speed and position information of integrated navigation,so that the tracking loop has low bandwidth.It still has high dynamic performance and higher accuracy of code phase estimation under multipath interference.The simulation results show that the auxiliary carrier loop can improve the dynamic adaptability and maintain high tracking accuracy.The auxiliary code loop can effectively suppress multipath interference and improve the robustness of the code loop.In this paper,the TC-OFDM/INS integrated navigation system is designed and implemented based on the hardware architecture of FPGA+ARM,and the indoor scene is tested and analyzed.The test results show that the positioning accuracy of TC-OFDM/INS integrated navigation system is 11.5%higher than that of TC-OFDM.It can still provide three-dimensional positioning results when the available base station signal is short. |
