Computationally Efficient Indoor Positioning Using the 3D-MUSIC Algorithm

In this paper, we propose a low-complexity solution for joint 3D super-resolution estimation of range, azimuth, and elevation angles to locate a transmitting user, using a 5G system with grid antenna arrays. The proposed approach leverages the 3D multiple signal classification (MUSIC) algorithm to accurately localize targets from a single temporal snapshot by smoothing the frequency-spatial covariance matrix. A significant contribution of this work is addressing the challenge of unknown time-of-departure, which complicates time-of-flight calculations. Our solution introduces two grid antenna arrays to mitigate this uncertainty, enabling precise localization even without perfect synchronization between the transmitter and receiver. Extensive simulations confirm the robustness and high accuracy of our method, demonstrating its strong potential for complex IoT applications in demanding industrial environments.