Research On Satellite Navigation Multipath Signal Simulation And Error Evaluation Method

In the outdoor complex environment,multipath effect is an important factor affecting the positioning accuracy of Global Navigation Satellite System(GNSS).With the rapid development of GNSS technology,the emerging high-precision navigation and positioning applications and algorithms have extensive anti-multipath testing requirements,which need to be tested under the multipath signal simulated by the navigation signal simulator or under the real signal to the sky.This paper focuses on multipath signal simulation and receiver positioning error evaluation method in real signal test.The specific contents are as follows:(1)Aiming at the lack of reasonable multipath signal parameters when simulating multipath signals,a complete GNSS deterministic multipath channel model including four modules: digital environment establishment,satellite and receiver trajectory generation,path search and electromagnetic field calculation is established based on image ray tracing.Aiming at the problems of large amount of calculation and low efficiency in the process of image ray tracing path search,a ray tracing acceleration method based on bidirectional back elimination is proposed,which greatly reduces the number of building faces to be processed in the environment.In the simulated urban canyon environment,the calculation time of this acceleration method is 92.08% less than that of the original ray tracing and 50.81% less than that of the unidirectional back elimination.The influence of building height on the characteristics of multipath signal is studied.It is concluded that the delay of multipath signal obeys gamma distribution,and the scale parameters are close to the mean value of multipath delay in the environment.(2)Aiming at the problem that the number of analog multipath signals in the existing navigation signal simulator is limited by channel resources and long operation time,a parallel simulation method of GNSS multipath signals based on GPU is proposed.The Farrow fractional delay filter is used to realize continuous and variable high-precision delay.On this basis,the steps of integer delay,amplitude attenuation and phase rotation are integrated,the independent parallel calculation model of multi-channel filter coefficients is established,and a large number of floating-point operation units in GPU are used for parallel calculation and filtering,the optimization strategies of data "blocking" calculation and asynchronous parallel execution are given.Computer simulation results show that the channel resource and time consumption of this method will not increase with the increase of the number of multipath signals.For the GNSS direct signal with sampling rate of 75 MHz and duration of 1 s,the maximum time-consuming for generating 50～1000 multipath signals using the proposed method is158.6ms.Compared with the serial simulation method on CPU,the proposed method speeds up to about 9.5 times,and any number of multipath signals generated by 6satellites can be simulated in real time.(3)Aiming at the problem that the previous positioning error evaluation has little correlation with the multipath environment of the receiver in the previous sky real signal test and evaluation system,a positioning error evaluation method based on the carrier coordinate system is proposed,which can decompose the positioning error vector into longitudinal error,transverse error and vertical error.The moving car experiment is carried out in the real environment.The results show that the surrounding environment of the road is closely related to the longitudinal error,transverse error.The longitudinal error of receiver positioning in the downhill scene of viaduct is significantly less than the transverse error.The multipath effect caused by high buildings in the scene of high buildings on one side causes the transverse error to be significantly greater than longitudinal error.Compared with the usual evaluation methods of positioning error in the East-North-Up coordinate system,the results show that the processing based on the carrier coordinate system can better separate the error and more accurately evaluate the influence of multipath effect caused by the surrounding environment on the positioning accuracy of the receiver.