基于张量分解的5G室分场景下的定位算法

Passive distributed antenna system (DAS) is expected to be used in 5G-based indoor positioning scenarios. However, the analog nature of DAS results in the received signal being a superposition of multiple antenna signals, posing significant challenges for accurate indoor positioning. In this paper, we proposed a tensor decomposition-based position estimation algorithm for 5G passive DAS system, which enhances positioning accuracy by effectively distinguishing the multiple antenna components. Specifically, we exploit the sparse characteristics of the received signal waveform to construct a low-rank tensor model, and then Block term decomposition (BTD) is employed to separate mixed multipath antenna components. Next, we estimate the time of arrival (TOA) based on the result of tensor decomposition. Finally, to eliminate the time synchronization error in TOA, we employ the time difference of arrival (TDOA) algorithm for location estimation. Simulation results show that our proposed indoor positioning method achieves accurate positioning performance with lower Computational complexity.