Robust TDOA Localization Using SVD Regularization Under Ill-Conditioned Receiver Configurations

doi:10.1109/tim.2026.3718121

Robust TDOA Localization Using SVD Regularization Under Ill-Conditioned Receiver Configurations

2026-07-29

0

PRE

AI

DOI：10.1109/tim.2026.3718121

Original

Original request for help

Save

Abstract

En 中文

Ill-conditioned sensor-target geometries in distributed time-difference-of-arrival (TDOA) measurement systems can severely degrade positioning accuracy under noisy conditions. This article proposes a robust singular value decomposition (SVD)-based regularized TDOA estimator to address this problem. By constructing a regularization matrix from the singular vectors corresponding to small singular values, the method suppresses the extreme variance amplification caused by ill-conditioned measurement matrices. A minimum mean-squared error (mse) criterion is applied to iteratively optimize the regularization parameters, and an analytic solution for the regularized TDOA estimate under SVD constraints is derived. The optimized parameters are then used to refine the measurement matrix, further enhancing measurement precision. Simulation and experimental results demonstrate that the proposed approach significantly reduces the RMSE and improves measurement stability compared with conventional TDOA and existing regularized least-squares methods. The method also introduces minimal estimation bias, providing a reliable and computationally efficient solution for accurate target sensing in practical instrumentation scenarios.

Keywords:

Ill-conditioned matrix

passive localization

regularization

singular value decomposition (SVD)

time-difference-of-arrival (TDOA)

Robust TDOA Localization Using SVD Regularization Under Ill-Conditioned Receiver Configurations

Abstract

Journal

Organization