A New Method for GNSS Multipath Mitigation with an Adaptive Frequency Domain Filter.

Siqi Yu,Xin Li,Renpan Wu,Fei Guo,Wanke Liu,Xiaohong Zhang

doi:10.3390/s18082514

Abstract

Multipath is the dominant error source for most fixed Global Navigation Satellite Systems (GNSS) sites and stations. The presence of multipath, particularly the multipath effect on pseudorange measurement, seriously affects positioning accuracy. Unfortunately, multipath effect reduction is still a challenging issue in high-accuracy GNSS positioning applications due to its special properties. To minimize the impact of the multipath effect, this paper focused on pseudorange multipath mitigation. First, the frequency spectrum of the code-minus-carrier divergence (CMCD) for Global Positioning System (GPS), Global Navigation Satellite System (GLONASS), and BeiDou Navigation Satellite System (BDS) satellites observed in different environments were analyzed, where we found that periodic fluctuations appeared in GPS and GLONASS as well as medium earth orbit (MEO) and inclined geosynchronous satellite orbit (IGSO) of BDS satellites in some situations, which manifested as peaks in the frequency domain. The results showed that the location of the frequency peaks in the frequency domain, width, and basic frequency spectrum intensity were different between different satellites and environments, causing difficulty in reducing the error impact. To eliminate such period fluctuations mainly caused by the multipath effect, a novel method based on a frequency domain filter was proposed in this paper. One of the keys of the proposed method was the use of short-time Fourier transformation (STFT) in the GNSS signal data processing to calculate the accurate local frequency spectrum when code-minus-carrier divergence (CMCD) was assumed to be time varying. Once the frequency spectrum was obtained, a new spectrum peak extraction method was used to locate the peak frequency position. By interpolation and inverse Fourier transformation, the influence of the spectrum peaks could be effectively eliminated, thus improving pseudorange precision. The experimental results showed that the periodic multipath effect could be greatly reduced by the proposed method.

Highlights

Various kinds of errors affect Global Navigation Satellite Systems (GNSS) positioning, such as ionospheric error, tropospheric error, orbital error, satellite and receiver clock errors, multipath error, and receiver noise
For Global Positioning System (GPS) satellites, most of the standard deviations decreased from 0.3~0.5 m to 0.2~0.3 m once the proposed method was applied for multipath mitigation
Despite all the effort put into mitigating multipath errors, it remains the dominant error source that cannot be ignored for GNSS precise positioning and other GNSS

Summary

Introduction

Various kinds of errors affect Global Navigation Satellite Systems (GNSS) positioning, such as ionospheric error, tropospheric error, orbital error, satellite and receiver clock errors, multipath error, and receiver noise. Among these items, multipath has been proven to be a limiting factor, which is difficult to be deal with. Surrounding objects generate many signal replicas that cause a large multipath disturbance in received signals, resulting in large positioning errors [1]. The characteristic related to the environment makes multipath difficult to model and correct. In most high-precision applications, multipath remains the dominate error source [2].

Methods

Results

Conclusion