Subarray-Based Hybrid-Field Channel Estimation for Terahertz Wideband UM-MIMO Systems Without Prior Location Knowledge
Yanran Sun, Chuang Yang, Mugen Peng
Abstract
Terahertz (THz) wideband ultra-massive multiple-input multiple-output (UM-MIMO) systems are critical for the upcoming 6G. The far- and near-field regions always coexist as the hybrid-field regions in the THz wideband UM-MIMO systems, but the existing hybrid-field estimation schemes can not perform well for the wideband channel without prior knowledge of user location information. In this work, we propose an efficient subarray-based hybrid-field channel estimation algorithm for the THz wideband UM-MIMO systems without prior knowledge of user location. The estimation problem is decomposed into a series of sub-problems, each of which considers only one subarray for channel estimation in combination with the wideband sparse detection. Then, the hybrid-field channel is recovered by combining all the subarray channels. Simulation results show that the proposed algorithm is able to recover the hybrid-field wideband channel accurately at a low pilot overhead and the bandwidth has little effect on the performance of the proposed.