Accurate GNSS Range Estimation in Multipath Environments Using Stochastic-Gradient-Based Adaptive Filtering

Multipath propagation is one of the dominant sources in ranging applications. The problem of GNSS multipath mitigation is investigated herein through the development of adaptive multipath channel compensation filters that impose a smaller computational load on the receiver compared to most of the advanced delay estimation techniques. An optimum estimation block is designed to produce the control error signal in the feedback loop used to update the filter coefficients adaptively. Using the proposed method, multipath distortion on the received signal is compensated for by passing signals through a filter whose coefficients are adaptively adjusted employing feedback signals. A set of simulation and real data processing results are presented to compare the performance of the proposed adaptive techniques with some of the conventional DLLs under realistic urban channel conditions. The experimental results are based on live vehicular GPS data collected in dense urban environments; the results demonstrate a reduction of up to 30% in RMSE values of the computed position errors.