Research On The Key Technologies Of Satellite Navigation Multipath Channel Modeling And Simulation

The optimization of the global navigation satellite system(GNSS)requires in-depth study of satellite navigation channel in typical propagation environments.With the development and construction of the high-medium-low-orbit joint satellite navigation system with a large number of satellites in the future,it will bring huge opportunities for satellite navigation multipath channel modeling and simulation,and it is also accompanied by higher expectations and more challenges.Therefore,the study of satellite navigation multipath channel modeling and simulation technology has important theoretical significance and application value for emerging navigation and positioning technology and new system navigation signal design in unsuitable environment.Although a lot of research has been done,there are still deficiencies and challenges: In terms of channel modeling,there are insufficient studies on satellite navigation time-varying channel modeling and GNSS navigation constellation multi-frequency channel modeling;in channel simulation,large-scale GNSS constellations and complex multipath brings serious challenges to realtime channel simulation.Based on the above background,this thesis conducts in-depth research on the key technical issues in satellite navigation channel modeling and simulation,starting from the two goals of making up for the lack of channel modeling research in typical scenarios and reducing the computational complexity of channel simulation.The main work and contributions of the thesis are summarized as follows:(1)For satellite navigation time-varying channel modeling,a channel modeling method based on multiple point-scatterers and kinematic model is proposed,which can accurately calculate the multipath continuous phase and time-varying Doppler shift of the channel,and provides an effective means to study the statistical characteristics of the time-varying channel.From the perspective of channel delay Doppler domain,the multiple pointscatterers model framework is introduced to construct the local dynamic environment of the mobile receiver,and the continuous time-varying motion velocity is mapped to the time-varying Doppler shift.The time-varying parameters of the channel are accurately updated by the efficient composite trapezoidal integration method,so as to track the instantaneous Doppler shift of the channel in the time-varying scene,reflecting the time-varying scattering effect fitting the motion state.Simulation results show that the proposed method can not only accurately model the time-varying characteristics of multipath continuous phase and Doppler shift,but also reproduce the statistical characteristics of time-varying channels.(2)For multi-frequency channel modeling of GNSS constellation,a multi-frequency hybrid channel modeling method combining ray-tracing and propagation graph is proposed.The GNSS channel model of urban canyon is constructed,and an effective modeling method of dense multipath channel is provided.From the channel time-frequency domain perspective,the advantage of hybrid modeling is introduced to reduce the complexity of modeling.Firstly,the ray-tracing method is used to model the channel deterministically,and the propagation graph method is further integrated to capture additional dense multipath components.A novel hybrid modeling method is formed,which can take into account the differences of multi-frequency propagation channels and obtain a multi-frequency channel model with high computational efficiency and spatial consistency.Simulation results show that this method can efficiently model the influence of dense multipath components.(3)For multipath channel simulation in delay domain,a multipath channel simulation method for wide delay range is proposed,and a multipath simulation model combining long and short delay is established to reduce the simulation complexity.From the perspective of channel delay domain,considering the long and short delay multipath distribution,different ways are adopted according to the characteristics of short delay and long delay multipath distribution,that is,the short delay multipath component is simulated by the tapped delay line simulation model,and the long delay multipath component is simulated by the discrete time approximation model.Under the condition of keeping the multipath error undistorted,the multipath delay can be extended in a wide range,and the resource consumption is not affected by the number of multipath,which reduces the complexity of channel delay domain simulation.The verification results show that the influence of the multipath channel simulated by the proposed method on the pseudo-noise code correlation peak of the navigation receiver is consistent with the original reference multipath channel.(4)For equivalent reduced multipath channel simulation,a channel sparse fitting method based on pre-clustering least-squares estimation is proposed,which can provide technical support for large-scale GNSS and complex multipath channel simulation modeling.This method is based on the idea of joint clustering learning and adaptive algorithm,that is,it is not only based on the clustering learning of multipath parameters in the delay power space,but also based on the least-squares estimation adaptive algorithm to minimize the error of the time autocorrelation function of the signal,which produces a more stable sparse fitting effect.The verification results show that this method has better performance than the traditional method,and has better practicability and robustness than the pure clustering learning method.It can decouple the number of taps in the tapped delay line structure from the number of multipath in the channel impulse response,which can greatly reduce the real-time computing resources and achieve the effect of equivalent simplified multipath channel simulation.In summary,the research results of this thesis make up for the lack of research on satellite navigation time-varying scenarios and GNSS constellation multi-frequency channel modeling,enrich the channel modeling methods and models in typical scenarios,and solve the reduction of multipath components in the delay domain.To solve the technical challenges of large-scale GNSS and complex multipath channel simulation,the simulation computing efficiency is further improved.