An Advanced Multipath Mitigation Method Based on Trend Surface Analysis

Zhiren Wang,Wen Chen,Yu Peng,Zhengqi Zheng,Danan Dong,Chenglong Zhang

doi:10.3390/rs12213601

Abstract

Among various ways to eliminate the multipath effect in high-precision global navigation satellite system positioning, the multipath hemispherical map (MHM) is a typical multipath correction method based on spatial domain repeatability, which is suitable for not only static environments, but also some dynamic carriers, such as ships and aircraft. So, it has notable advantages and is widely used. The MHM method divides the sky into grids according to the azimuth and elevation angles of satellite, and calculates the average of the residuals within the grid points as its multipath calibration value. It is easy to implement, but it will inevitably lead to excessive or insufficient multipath correction in the grid. The trend surface analysis-based multipath hemispherical map (T-MHM) method makes up for this deficiency by performing trend surface analysis on the multipath spatial changes within the grid points. However, the effectiveness of T-MHM is limited and less capable of resisting noise interference due to the multicollinearity between the independent variables caused by the special spatial distribution of multipath sampling and the overfitting problem caused by ignoring the multipath anisotropy. Thus, we proposed an improved multipath elimination method named AT-MHM (advanced trend surface analysis-based multipath hemispherical model), which cautiously judges the occurrence of the above problems and gives corresponding solutions. This was extended to double-difference mode, which expands the scope of application. The performance of AT-MHM in GPS pseudorange multipath mitigation was verified on geodetic receiver and low-cost receiver in a strong multipath environment with high occlusion.

Highlights

Multipath error is one of the key unresolved errors in global navigation satellite system (GNSS) positioning, as it is difficult to parameterize and cannot be eliminated by differential technology
The results show that the AT-multipath hemispherical map (MHM) model achieved the best results in three directions, and the baseline accuracy was improved by approximately 45.98%, 58.68%, and 43.93% in the three directions of east, north, and vertical, respectively (Table 5)
The multipath mitigation methods based on multipath spatial repeatability have the advantage of being able to be applied to static or dynamic environments in real-time correction

Summary

Introduction

Multipath error is one of the key unresolved errors in global navigation satellite system (GNSS) positioning, as it is difficult to parameterize and cannot be eliminated by differential technology. The methods to suppress multipath error are mainly divided into hardware-based and software-based. The hardware-based method includes improvements of the antenna and receiver. The antenna improvement methods mainly include the patch elements placed on choke rings [1], the multipath estimating delay-locked loop [2,3] and a “narrow correlator” delay-locked loop [4], which can only partially eliminate multipath error. Antenna-related improvements include array antennas [5,6,7,8,9,10] and dual-polarized antenna techniques [11,12,13]. The above methods effectively suppress the multipath effect, but increase the hardware cost

Objectives

Methods

Findings

Discussion

Conclusion