Multipath Error Estimation Based On Statistical Model And Multipath Mitigation Technique For Signals Of Modern Global Navigation Satellite Systems

The multipath signal in the Global Navigation Satellite System(GNSS)has degraded the code and carrier measuring accuracy,and it is highly correlated with the surroundings,antenna characteristics,signal modulation and receiving algorithms.Multipath effect is considered as one of the major error contributor because it is difficult to be cancelled by differential approach.This thesis studys the code multipath mitigation methods based on the multipath error estimation,BOC signal multipath mitigation and static multipath error,including four contributions as follows:i)A modified multipath error assessment metric based on the multipath signal model is proposed to solve the irrelevance between the traditiona multipath error assessment metric and reciving environment.Two indicators of modified multipath error envelope and average multipath error are proposed to evaluate the receiver multipath mitigation performance under certain circumstance.The research shows the average error of BPSK(1)signal is improved by 5%~15% with BOC(1,1)signal instead.Two typical receiving environment with multipath parameters are proposed for the multipath estimation methods.The performance of multipath esitmation delay lock loop and Teager-Kaiser algorithm are analyzed,and they are found to be working badly with multiple middle-delayed strong multipath signal.ii)A least-squares approximation method with a ideal target discriminator is proposed,utilizing the truncated singular value decomposition method.It aims at solving false lock problem in the code tracking of BOC signals.By this method,local correlators and code correlation reference waveforms can be designed with no ambiguities in the discriminator curves.The results show that the proposed method can realize coherent and non-coherent BOC signal discriminators without false lock points.Generally,there is performance degradation in the tracking jitter.However,there are performance promotions in the multipath error envelope for the BOC(1,1)signal with 77.62% area decrease,and 3 to 5 dB increase in the tracking precision compared with W2 waveform.iii)To solve the multipath mitigation problems of high-order BOC signal unambiguous tracking method and to increase the tracking precision of the least-squares approximation method,the double phase estimator(DPE)was modified by introducing a strobe waveform in the prompt signal correlation process of the subcarrier phase lock loop(SPLL)integration.Moreover,a technique of DLL discriminator combined with TK operator is proposed.Simulation results show that,relative to conventional DPE,the first waveform employed in this study provides a reduction in the subcarrier multipath error envelope(SMEE)area of 81.1% and 75.1% for BOC(1,1)and BOC(14,2)signal respectively.The second waveform employed in this study provides a reduction in the SMEE area of 82.5% and 76.8% for BOC(1,1)and BOC(14,2)signal respectively.They also outperforms the double estimator(DE)about 64.4% and 53.2% for BOC(1,1)and BOC(14,2)signal in the SMEE area.The multipath mitigation performance of TK operator is enhanced by more than 10 dB for BPSK(1)signal in the city environment.The multipath error envelope of BOC(1,1)signal with new technique is 14.8% of the W2 method with 24.552 MHz bandwidth.iv)A frequency variation approach for geostationary earth orbit(GEO)satellite standing multipath mitigation,which can effectively enhance the traditional filtering method performance,is proposed.This method aims for decreasing the GEO slowly multipath ranging error,which cannot be mitigated by data processing method.Simulation results show the root mean square error(RMSE)curve of proposed approach can reach 74.7%~92.6% improvements in GEO satellite multipath mitigation with 40 MHz frequency variation range.A non-integer carrier phase ambiguity solving method is proposed for the carrier varying signal receiving.Also,a truncated singular value decomposition method is proposed for the parameter matrix morbidity problem in the integer ambiguity resolution with short epoches.Finally,the conclusions and research content are drawn and discussed in Chapter VI.The application of research results in the projects are presented,and the the future work that can be done based on this thesis are indicated.