Two-step unbiased estimation on small-delay multipath of GNSS signals based on S-curve bias

• Main Authors: Jin, Tian, Zhang, Pei, Yuan, Heliang, Cong, Li, Lu, Xueyan, Qu, Jiaqing
• Other Authors: School of Electrical and Electronic Engineering
• Format: Article
• Language: English
• Published: 2023
• Subjects: Engineering::Electrical and electronic engineering; Global Navigation Satellite Systems; Small-Delay Multipath
• Online Access: https://hdl.handle.net/10356/170747
• Institution: Nanyang Technological University

Multipath error is among the main error sources of global navigation satellite systems (GNSSs). The traditional estimation method represented by the multipath estimating delay-locked loop (MEDLL) has difficulty in distinguishing small-delay multipath signals and noise during iterations, which leads to a bias in the estimation of small-delay multipath signals. To mitigate this small-delay multipath estimation bias, a two-step unbiased estimation method based on the S-curve bias (SCB) is proposed. First, the relationship among SCB, multipath parameters, and correlation spacing is derived. Second, a search approach based on the least squares is designed to obtain the turning point position of the SCB results with different correlation spacings. Third, the multipath delay can be estimated through the turning point of the SCB results. Finally, the phase and amplitude ratio of the direct signal and multipath signals is estimated by the SCB results of local carrier signal replicas with different phases. The proposed method does not contain iteration steps, which minimizes the accuracy influence introduced by the iterations. Simulations using the Spirent simulator and real signal experiment results show that the proposed algorithm can reduce the small-delay multipath error effectively with low computational complexity. Compared with MEDLL, Teager and Kaiser-MEDLL (TK-MEDLL), and high-resolution correlator (HRC) algorithms, the small-delay multipath mitigation performance of the proposed algorithm was improved by approximately 82.9%, 74.5%, and 91.7%, respectively. Therefore, the proposed algorithm can mitigate estimation bias in small-delay multipath signals effectively. This method provides a new strategy for the parametric estimation of the multipath signals of GNSS.