Interference Suppression And Multipath Mitigaition Methods For GNSS Receivers

Global Navigation Satellite System (GNSS) can provide highly accurate globalpositioning, navigation and timing services. But the received signal power is very low,which is susceptible to intentional and unintentional interferences, resulting inpositioning performance loss and even failed to locate. The multipath signal is also amajor source of GNSS positioning error. Adaptive arrays is one of the importantmethods for GNSS interference suppression. However the GNSS signal power is suchlow that the steering vector is hard to estimate accurately, resulting in the performancedegradation of the classic beamforming algorithms. In order to tackle this problem, thebeamforming algorithm which does not rely on the signal steering vector is needed. Inorder to obtain a better interference suppression performance of the GNSS receivers, thespace-time adaptive processing (STAP) is introduced. Unfortunately, the time domainfiltering structure distorts the GNSS signals, thus the time-domain response of theweight vector should be well designed. Multipath signals cause severe positioning error,the receiver needs to suppress multipath signals. This dissertation focuses on theproblems which are encounted in GNSS applications, involving the interferencesuppression, space-time adaptive processing effects on GNSS signals and multipathmitigation. The main contribution of the work are listed as follows:1. The accurate steering vector of the GNSS signals which is needed by adaptivealgorithms for interference suppression is hard to obtian. In order to tackle this problem,a novel interference suppression algorithm for GNSS signal is proposed. The spreadingcodes in GNSS signal exhibit strong self-coherence and the correlation matrix of theGNSS signal can be estimated using several peaks of correlation values of receivedsignal and reference signal. The optimum weight is obtained under the maximum SINRcriterion using the covariance matrix of pre-correlation and post-correlation signal. Theproposed algorithm does not need a priori knowledge of the directions of GNSS signals.Simulations and experiments demonstrate that, compared with the power inversionmethod and SCORE algorithm, the proposed algorithm can provide a similarperformance of inference suppression, but the output SINR is significantly increased.2. The correlation function of the GNSS signals and the reference signal is theinput signals of all processing in GNSS receiver. In order to study the effect of adaptivearray processing on GNSS signals, the correlation function of the output GNSS signalsof spatial adaptive and STAP are derived respectively. The effects of adaptive arrayprocessing on GNSS code tracking loop and carrier tracking loop are analyzed based on the derived correlation funcion. The analysis results show that adaptive spatialprocessing has no effects on GNSS signal, while STAP distorts the GNSS signal due tothe lack of distortion constraint of GNSS signal waveform, resulting in positioningperformance loss. As the distortion of GNSS signals is caused by the time-domainfiltering structures, the effects of delay tap and the weight vector of the space-timeadaptive processing are analyzed respectively. An optimized structure which canmitigate the distortion is proposed. Finally conclusion is verified by simulation. A linearphase space-time adaptive algorithm for interference suppression is proposed in order tomitigate the distortion of GNSS signal which is induced by STAP. First the equivalentcomplex FIR filter of the STAP is derived, which can accurately describe the impact ofspace-time adaptive processing on GNSS signals. By constraining the time domainresponse of the equivalent filter so as to satisfy linear phase conditions, an optimizationmodel of linear phase space-time adaptive algorithm is obtained. The constraints istransformed and combined to obtain a closed-form solution. The computationalcomplexity of the proposed algorithm is similar to that of the classic algorithms.Simulation results show that the proposed algorithm has a similar interferencesuppression capability with the classic space-time adaptive algorithm, while the codetracking performance and carrier tracking performance of GNSS signals aresignificantly improved compared with the classic space-time adaptive algorithms.3A novel technique for mitigating the multipath-induced code delay estimationerror in GNSS is proposed. In contrast to conventional methods that aim to eliminatemultipath signals, the proposed method exploits them to enhance the direct signalwithout affecting the accuracy of GNSS code delay estimates. To achieve this, coherentaccumulation of the received GNSS signals is first done by transforming the receiveddata into frequency domain and the parameters of multipath signals are then estimatedby sparse reconstruction algorithm. Subsequently, a modified local reference signal isemployed in delay lock loop (DLL) of the GNSS receiver, which mitigates thepseudo-range estimation error and increases the correlation value of direct GNSS signal.Simulation results demonstrate the performance and robustness of the proposed method.