Answers are grounded in this paper and cite their sources.
Long-baseline (LBL) time difference of arrival (TDOA) localization in shallow water often suffers from degraded performance when the sound speed is unknown. Most methods typically assume a single global equivalent sound velocity (ESV) shared by all propagation paths, which neglects the spatial variability of ESV and leads to degraded localization accuracy. Modeling the ESV between the target and each receiving sensor as an independent variable can more accurately represent the propagation characteristics, but it also introduces a new challenge: how to simultaneously estimate and update multiple path-dependent ESVs while maintaining algorithmic robustness and stability. To address this issue, this study proposes a multi-constrained least squares (MCLS) framework that jointly estimates the ESVs and the target position. Specifically, the MCLS algorithm consists of a three-stage procedure. First, an alternating direction method of multipliers (ADMM) based module estimates a unified ESV under multiple constraints to provide a reasonable initialization for subsequent localization. Second, the updated ESV is incorporated into a separately constrained weighted least squares (SCWLS) framework to obtain a coarse position estimate. Finally, another ADMM stage jointly refines the position and path-dependent ESVs. Simulation and experimental results demonstrate that the proposed method achieves robust and accurate localization under unknown sound speed conditions, outperforming existing methods.
No takes yet. Share an insight, caveat, or question.
Liu et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: