TDOA-Based Indoor Localization via Linear Fusion With Low-Rank Matrix Approximation

Abstract

Target indoor localization has become an attractive research topic due to its importance in location-based applications in wireless networks for sensing, controlling, and communicating. In TDOA-based localization models, timestamp packets must be exchanged between anchor nodes and target nodes. Timestamp measurements are susceptible to random transmission delays and packet loss in indoor environments, resulting in inaccurate positioning accuracy. In this paper, we propose a linear fusion indoor localization scheme based on TDOA with low-rank approximation to improve target localization accuracy and robustness. In an asynchronous localization model, we first formulate the indoor localization problem from incomplete and noisy timestamp measurements as a low-rank matrix completion problem. Furthermore, the proposed linear fusion algorithm is used to further optimize the localization accuracy by weighting multiple localization rounds. Simulation and experimental results indicate that the proposed method is more effective than the existing methods in the presence of packet loss and random delays.