Time-Delay estimation via CPD-GEVD applied to tensor-based GNSS arrays with errors

Abstract

Safety-critical applications (SCA), such as autonomous driving, and liability critical applications (LCA), such as fisheries management, require a robust positioning system in demanding signal environments with coherent multipath while ensuring reasonably low complexity. In this context, antenna array-based Global Navigation Satellite Systems (GNSS) receivers with array signal processing schemes allow the spatial separation of line-of-sight (LOS) from multipath components. In real-world scenarios array imperfections alter the expected array response, resulting in parameter estimation and filtering errors. In this paper, we propose an approach to time-delay estimation for a tensor-based GNSS receiver that mitigates the effect of multipath components while also being robust against array imperfections. This approach is based on the Canonical Polyadic Decomposition by a Generalized Eigenvalue Decomposition (GPD-GEVD) to recover the signal for each impinging component. Our scheme outperforms both the Higher-Order Singular Value Decomposition (HOSVD) eigenfilter and Direction of Arrival and Khatri-Rao factorization (DoA/KRF) approaches, which are state-of-the-art tensor-based schemes for time-delay estimation, particularly when array imperfections are present.