Carrier Tracking Technology For Weak GNSS Signals

Global Navigation Satellite System(GNSS)has high precision and can be used all day and in all weather conditions.So it has been used widely in civil and military fields.Because the orbit height of GNSS satellites are high,so the propagation loss is great.The signal intensity on the ground only can reach to-135~-130 dBm.In some harsh environments,such as urban area,forest and indoor,the signal will be further attenuated 10~30dB.This brings a serious challenge to the GNSS receivers.Especially the phase-locked loop(PLL),which is the weakest link in GNSS receivers,will output big errors and even lose lock in these environments.The carrier tracking loop is one of the most important links in GNSS receivers.Its precision will cause velocity errors and influence the precision of the code phase-locked look.So the performance of the positioning and navigation will decrease seriously.The carrier tracking in weak signal environments is always one of the difficulties which limits the application of GNSS greatly.Focus on the weak signal environments such as indoor and urban,this paper is aimed at improving the carrier tracking sensitivity,precision and robustness.Based on the coherent integration results,this paper proposes some lowcomplexity and high-performance carrier tracking loops to be used in different environments and for different signals.The performance of some loops are verified by a semi-physical experiment system.The main work includes the following sections.1.A carrier tracking loop using GPS(Global Positioning System)L1C/A signals is proposed for weak signal circumstances.This loop consists of an maximum likelihood estimator(MLE)and an adaptive Kalman filter(AKF).ML estimator estimates the carrier parameters including the amplitude,frequency and phase using the coherent integration results.Then the estimation results are sent to the AKF to get the filtering results.Because the coherent integration results are processed,the amount of data is small.What's more,an efficient search algorithm is used to search the optimal solution,this estimator has much lower complexity than the ML-based methods and can reach the Cramer-Rao lower bound(CRLB).The AKF can smooth the parameter estimation and get higher precision.The simulation results show that the proposed loop can get higher tracking sensitivity,precision and lower bit error rate.And this loop is more suitable for low-dynamic circumstances.2.A carrier tracking loop using GPS L5 pilot channel is proposed for weak signal circumstances.This loop consists of an non-linear least squares(NLS)estimator and an AKF.The first order Taylor series expansion of the result is made in the NLS estimator first.Then the carrier parameters are estimated by linear least squares.Because the first order Taylor series expansion introduce nonlinear error,NLS estimator is not an optimal estimator.But compared with ML estimator,NLS estimator doesn't need to search the optimal solution iteratively,so it has lower computation burden.The simulation results show that this loop is sensitive to the data bit reversal and not suitable to the signal with data modulation.But for the signal without data modulation,such as Global Positioning System(GPS)L5 pilot channel,this loop can reach higher performance than the ML estimator.3.A carrier tracking loop is proposed which is suitable for dynamic and weak signal circumstances.In dynamic and weak signal environments,the carrier tracking loop based on Fast Fourier Transform(FFT)is seriously influenced by the acceleration and limited by the resolution.To solve this problem,this paper proposes a 2-step tracking loop based on discrete chirp Fourier transform(DCFT)and phase fitting.This loop estimates the carrier Doppler frequency and Doppler rate roughly by DCFT first.Then the precise estimation of the carrier parameters such as carrier phase,Doppler frequency and Doppler rate is got by phase fitting.After introducing the phase fitting,the computation burden of DCFT is reduced and the resolution problem is solved too.Because the 2-step loop can estimate the Doppler rate accurately,the effect of acceleration is reduced greatly.The simulation results show the proposed method can improve the tracking sensitivity and precision than the traditional FFTbased and DCFT-based loops in dynamic and weak signal circumstances.4.A carrier tracking loop is proposed which is suitable for urban circumstances.The signal tracking in dense urban area faces some problems including weak signal,multipath interference and signal interruption.To improve the carrier tracking sensitivity and robustness in urban area,a tracking loop which joints all the channel data is proposed.Different from the traditional loops which process the data in each channel,this loop fuses the data of all the channels to build the cost function of the receiver velocity.Then the velocity estimation can be obtained by searching the maximum value of the cost function.So this loop achieves the data fusion in velocity domain.It can reach higher signal-to-noise ratio(SNR)gain,solve the signal interruption problem in one channel and has higher resistance to multipath signals.The simulation results show that the proposed loop can improve the tracking sensitivity and has higher robustness to the signal interruption and multipath interference.It is suitable for the navigation in urban area.Through the study in this paper,the main problems in some weak circumstances such as urban and indoor are pointed out.By processing the coherent integration results to achieve the low complexity,this paper completes a high-sensitivity carrier tracking loop based MLE and AKF,a carrier tracking loop based NLS and AKF for the pilot channel,a carrier tracking loop based on DCFT and phase fitting in dynamic and weak signal,a carrier tracking loop which joints all the channel for urban navigation.The theory analysis and simulation results proves these loops have higher performance than the traditional loops.These achievements complement and perfect the tracking theory in weak GNSS signal environments.They provide the theoretical supplement for the design of the high-sensitivity GNSS software receiver too.