Study For Processing Weak Signal With Multipath Effect And Positioning On Software Defined GNSS Receiver

Global Satellite Navigation System(GNSS)is restricted in applying the environments such as forests,urban valleies.The interference to GNSS signals is serious,which leads to weak signals and signals with multiple arrivals.Due to the poor performance of acquisition and tracking caused by the weak signal and multipath effect,errors in the estimate of pseudorange and the final position.In this paper,the research of processing weak signal with multipath effect and positioning on software defined GNSS receiver is carried out,which aims to improve the performance of GNSS acquisition,tracking and positioning.Main research contents and achievement are as follows:1 The acquisition and processing system of satellite data based on USRP was built according to the GNSS signal characteristics and positioning principle,where acquisition,tracking and positioning are all implemented in the software defined GNSS receiver.Due to the fact that,in USRP,the sample frequency cannot be generated as the integer multiple bit rate of the pseudorandom noise(PRN)code,which lead to additional code phase error during the tracking step,a pseudorange estimation method is further presented to improve the positioning accuracy.2 The method to process weak signal based on parallel code phase search was studied,which aims to improve the ability to acquire weak signals.Since the self coherence based on C/A code,signal accumulation in the form of combining coherent integration with non-coherent integration was to enhance the signal recognition.Decision mode of acquisition was optimized by taking the ratio of the peak to the average in a range as the decision variable.The Fourier transform was used to refine the frequency,which build the relationship between sampling points and frequency3 Three methods of mitigating multipath effect were studied,which include Early-Minus-Late Loop(EML),Early late slope technique(ELS)and High Resolution Correlator(HRC).For the deficiency of HRC in mitigating multipath effect,an optimized method to process multipath signal based on HRC was proposed,which combines the technique of narrow Early-Minus-Late Loop.The influence of the spacing which should be chosen reasonably on the error estimation was considered.4 Four positioning methods were studied,i.g.,nonlinear iteration method,line of position(LOP)method,semidefinite programming(SDP)method,constrained weighted least squares by introducing an extra variable,which were applied in the GPS positioning.The model of simulation positioning was constructed to analyze the performance of each method.To mitigate the error of pseudorange measurement and the nonlinear problem caused by weak signal and multipath effect,solutions were further presented to improve the positioning accuracy,which include e.g.regularization modification of Gauss Newton method,linear modification or separate operation of weighted constrained least squares method,error correction of two step method.The results of simulation and field tests show that the proposed method can effectively improve the ability of processing multipath and weak signals,and reduce the generation of positioning errors.Under the multi-path and weak signal environment of field text,the positioning error of software defined GNSS receiver is 4.08 m,which is based on the method of acquisition,tracking and positioning in this paper.