Research On The COMPASS Precise Positioning Algorithms And Multipath Effects

BeiDou Navigation Satellite System(COMPASS)is a navigation satellite system totally developed by Chinaindependently.Its function will gradually transition from regional navigation positioning system to global navigation positioning system followed by the overall development roadmap of “regional service first,global service later”.As the COMPASS is in constant development and improvement stage,it will play a important role to improve the system application scope that research on the key techniques of the COMPASS precise positioning.In addition,the multipath error has become an important factor of affecting the positioning accuracy due to it can't be eliminated by differential technique in the COMPASS kinematic positioning.Therefore,the thesis has mainly made a research on the COMPASS precise positioning algorithms and multipath effects based on the COMPASS positioning principle.The main works are summarized as follows:For BOC modulation mode adopted by the COMPASS navigation signal,firstly introduces in detail the principle and characteristics of BOC modulation signal.The thesis mainly introduce the pseudorange positioning observation model and carrier phase observation model,and the single difference,double difference and triple difference model based on the pseudorange and carrier phase observation.Finally,the measurement errors of the COMPASS positioning have been systematically analyzed.MBOC(6,1,1/11)modulation signal is adopted by the COMPASS B1 spectrum civial signal.In view of the existing MBOC signal tracking ambiguity problem,a MBOC signal ambiguity elimination technique based on the pseudo correlation function(PCF)is proposed.The proposed technique firstly designs two local auxiliary signals,correlates with the received MBOC signal respectively,and then constructs a unambiguous pseudo correlation function.Simulation results show the PCF technique can completely eliminate the side-peaks of MBOC signal and achieve unambiguous MBOC signal tracking.According to the characteristic of MBOC signal autocorrelation function,a MBOC signal side-peaks cancellation technique by combining the sub-correlation function(CSCF)is proposed.The MBOC signal autocorrelation function is decomposed into sub-correlation functions,it can obtain the correlation function without side-peaks by combining the sub-correlation functions.Simulation results also show that the CSCF technique can completely eliminate side-peaks of MBOC signal and achieve unambiguous MBOC signal tracking.The traditional principle of LAMBDA algorithm is analyzed when the carrier phase measurement is used to solve.Decorrelation processing is the most critical step in theLAMBDA algorithm.Aiming at the disadvantage and limitation of LLL algorithm,a modified LLL algorithm based on the row descending adjustment matrix and modified Gram-Schmidt orthogonal transform is proposed.It can improve the decorrelation effect of LLL algorithm and is more suitable for processing the high dimensional integer ambiguity covariance matrix.An integer ambiguity fast resolution algorithm based on adaptive genetic algorithm is proposed to improve the solution efficiency of integer ambiguity.Float solution can be obtained by using the integer least square.The adaptive genetic algorithm is introduced to search integer ambiguity after coding,determing the fitness function,designing appropriate genetic operators and operating parameters.Simulation results show that the search efficiency of algorithm is superior to LAMBDA algorithm.When kinematic navigation data processing,the parameter estimation model of kinematic relative positioning is firstly introduced.Then a adaptive robust filter algorithm with classification factors based on the bifactor weight and windowing method is proposed.In this algorithm,the state parameters are divided into position parameters and velocity parameters,the observation vector equivalent matrix is determined by designing appropriate bifactor,the state prediction equivalent matrix is determined using the principle of windowing fitting method.Simulation results also show that the algorithm can sffectively control the impact of system disturbance and deviation motion model to parameter estimation.The multipath error is always one of the important error sources in COMPASS positioning.Firstly,the COMPASS signal with MBOC modulation is systematicly analyzed in multipath environment.According to the autocorrelation function of MBOC signal,the mathematical model of multipath error envelope and running average multipath error envelope are established to obtain the influence of multipath amplitude attenuation coefficient,the front-end bandwidth,the correlator spacing to code tracking error and the influence of the amplitude attenuation factor,the carrier phase relative difference to the carrier phase error.The thesis proposes a multipath estimation algorithm based on the multi-correlator COMPASS receiver.Firstly,determine the number of correlators,use the maximum likelihood estimation and grid search method to obtain the maximum likelihood estimation value of the signal,and then use the piecewise linear interpolation method to obtain the ling-of-signal(LOS)delay calculation,so as to achieve more accurate LOS estimation value.