Design And Performance Evaluation Of Tracking Loop For Satellite Navigation Receiver

Global navigation satellite system(GNSS)plays an increasingly prominent role in social development and military application.It is an important support of national economy and a key embodiment of national comprehensive scientific and technological strength.With the progress of society and the complexity of application environment,people’s demand for navigation and positioning is increasing,and the performance demand of satellite navigation receiver is also increasing.In GNSS,the acquisition and tracking of signals by navigation receiver will directly affect the final navigation output accuracy and determine the performance of GNSS receiver.The principle of traditional GNSS tracking loop is scalar tracking.The measurement of code phase and carrier frequency of each tracking channel is independent of each other,and the common information between tracking loops is not fully explored.As one of the research hotspots of modern navigation and tracking technology,vector tracking can fuse all tracking channel information and combine with positioning output,which makes GNSS receiver have the ability of tracking weak signal,high dynamic signal and anti-interference without any external hardware assistance.In addition,as an important part of GNSS,pseudolite system has shown incomparable technical advantages in GNSS signal occlusion environment.Specifically,pseudolite system can further improve the positioning accuracy,integrity and availability of GNSS system.Different from GPS and other systems,there is no strict clock synchronization mechanism in the pseudolite system.The clock offset between pseudolites will lead to a new tracking error in the receiver tracking loop,which will seriously affect the performance of navigation receiver.Therefore,the tracking loop design of navigation receiver for pseudolite signal is also the key technology of signal processing.In view of the application background of satellite navigation system and the development trend of receiver design,this paper conducts research on the tracking loop design and performance evaluation of GNSS and pseudo-satellites,aiming at the problems such as low GNSS vector tracking robustness and large clock deviation impact of tracking loop error of pseudoranets.The main contents and obtained research results are as follows:(1)The performance of vector tracking technology is evaluated.By fusing the information of all satellite channels and combining with navigation solutions,vector tracking technology continuously optimizes the estimation of user position,speed,clock offset and clock drift.Compared with the traditional scalar tracking technology,vector tracking has the advantages of bridging instantaneous signal,resisting multipath and tracking high dynamic and weak signals.But scalar tracking and vector tracking have different noise bandwidth characteristics.In view of this,from the perspective of equivalent noise bandwidth,this paper uses simulation experiments to evaluate the performance of vector tracking in high dynamic environment.The simulation results show that the performance of vector tracking loop is better than that of traditional scalar tracking even under the condition of equal noise bandwidth.The reason is that vector tracking can combine each tracking channel,so that the channels can assist each other and improve the tracking performance.(2)The noise parameter estimation of vector tracking algorithm is studied.The core of the existing vector tracking loop is extended Kalman filter(EKF),whose filtering performance depends on the precise setting of system model and noise statistical characteristics.In practical application,it is difficult to set the noise item accurately with the change of environment.This makes the filtering accuracy greatly reduced,and even leads to filter divergence,which seriously affects the tracking accuracy of the receiver.Aiming at the problem of noise estimation in vector tracking,this paper proposes an adaptive noise estimation method based on neural network.By analyzing the noise error correlation in the tracking process,the loop tracking parameters and the expected observed noise covariance are selected as the basis for network training.Neural network is used to mine the internal relationship between the tracking parameters and the covariance of observation noise.Through experiments,the effectiveness of the proposed method is verified in the case of high carrier to noise ratio and low carrier to noise ratio,and compared with the traditional method.The results show that the tracking and positioning accuracy of the proposed method is improved compared with the traditional method,which proves the effectiveness and feasibility of the proposed method.(3)The robustness of vector tracking algorithm is studied.A vector tracking loop design method based on unbiased finite impulse response(UFIR)filter is proposed.Firstly,the principle of UFIR filtering algorithm is introduced,and the characteristics of the algorithm are analyzed and compared with EKF,and the ufir vector tracking loop model is established.Secondly,in order to further improve the tracking performance,a fusion filter based on Gaussian clustering principle is designed.The filter takes EKF and ufir as sub filters and provides fusion mechanism based on Gaussian clustering criterion.Finally,a new method is proposed to improve the tracking performance.The experimental results show that the proposed method can adjust the weights of the two filters adaptively according to the noise statistics,and the filter can take into account the estimation accuracy of EKF and the robustness of UFIR algorithm.(4)The influence of clock error on tracking loop error in asynchronous pseudolite system is studied.In order to realize high precision positioning of pseudolite system,the clock offset between pseudolites must be eliminated.The existing methods mainly use the reference receiver to build carrier phase double difference algorithm to solve the problem.But the precondition is that the clock synchronization between user receiver and reference receiver is required.This paper focuses on the impact of clock difference between receivers on tracking error and positioning performance in asynchronous pseudolite system.Based on the self-developed prototype of pseudolite system,the theoretical derivation and experimental verification are analyzed respectively.The results show that due to the clock difference between the user receiver and the reference receiver,the double difference method can not completely eliminate the frequency deviation between pseudolites.Finally,two solutions based on hardware and software are proposed to solve the problem.Experiments show the effectiveness of the proposed method.