Direct Position Determination with Distributed Sensor Arrays Under Multi-path Environment: Matrix Reconstruction and Subspace Data Fusion

Subspace-based direct position determination (DPD) using distributed sensor arrays is challenged by rank-loss of the noise-free data covariance matrix in a multi-path environment. Spatial smoothing methods are commonly utilised to address this challenge by restoring the matrix rank. Nevertheless, traditional approaches such as forward spatial smoothing pre-processing (FSSP), modified spatial smoothing pre-processing (MSSP), and enhanced spatial smoothing preprocessing (ESSP) can result in aperture loss in sensor arrays. This article presents a novel DPD method employing matrix reconstruction and subspace data fusion (SDF). The proposed method effectively mitigates rank loss in the data covariance matrix without compromising sensor array aperture, thereby enhancing localization accuracy compared to conventional methods. The efficacy and superiority of the proposed method are validated through Cramér-Rao Bound (CRB) analysis and simulation results.