Error Ellipse-Based Outlier Detection for GNSS Doppler Measurements in Urban Environments

Taeho Kim, Jeonghyeon Yun, Byungwoon Park

Abstract:	In urban environments, Global Navigation Satellite System (GNSS) positioning suffers from severe accuracy degradation due to multipath and non-line-of-sight (NLOS) signals. This performance deterioration becomes even more pronounced in smartphone GNSS, which is characterized by low-cost antenna structures and irregular signal-to-noise ratio (SNR) characteristics. To address these limitations, Doppler-based velocity estimation techniques have been proposed; however, in dense urban settings, Doppler measurements are also contaminated by reflections and non-line-of-sight (NLOS) signals, resulting in numerous outliers. The Least Squares (LS) method, which minimizes the sum of squared residuals for all observations, is highly sensitive to such outliers, and the presence of extreme measurements can easily distort the estimation results. Iteratively Reweighted Least Squares (IRLS) offers improved robustness against a limited number of outliers but inevitably degrades in performance when many outliers are present. Similarly, Receiver Autonomous Integrity Monitoring (RAIM) assumes only a small number of faults and is therefore unsuitable for urban scenarios. Although Random Sample Consensus (RANSAC) has been introduced as an outlier elimination technique, it can incorrectly identify subsets containing outliers as the consensus set in urban environments, leading to biased model estimation. To overcome these limitations, this study proposes an enhanced RANSAC algorithm that incorporates an Error Ellipse -based Subset Consistency Check. By verifying the distributional consistency of subsets under the assumption that inliers form a homogeneous distribution, the proposed method mitigates the risk of selecting outlier subsets as the consensus. Experimental results in urban environments demonstrated that the proposed algorithm improved upon LS, IRLS, and conventional RANSAC in terms of both accuracy and reliability and enhanced the accuracy of velocity estimation in urban scenarios.
Published in:	Proceedings of the 38th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2025) September 8 - 12, 2025 Hilton Baltimore Inner Harbor Baltimore, Maryland
Pages:	1112 - 1119
Cite this article:	Kim, Taeho, Yun, Jeonghyeon, Park, Byungwoon, "Error Ellipse-Based Outlier Detection for GNSS Doppler Measurements in Urban Environments," Proceedings of the 38th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2025), Baltimore, Maryland, September 2025, pp. 1112-1119. https://doi.org/10.33012/2025.20317
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In