Channel Estimation in Radiative Near Field Region of 6G Extremely Large MIMO System

• Published in: Advances in electrical and computer engineering Vol. 24; no. 3; pp. 69 - 76
• Main Authors: SUGANYA, T., INDIRA GANDHI, S.
• Format: Journal Article
• Language: English
• Published: Suceava Stefan cel Mare University of Suceava 01-08-2024
• Subjects: 6G mobile communication; Accuracy; Algorithms; Antenna arrays; Antennas; Communications systems; Domains; Error analysis; Far fields; Matched pursuit; Near fields; Signal to noise ratio; Simulation; Sparsity; Spherical waves; Wireless communications
• ISSN: 1582-7445; 1844-7600
• DOI: 10.4316/AECE.2024.03007

6G wireless communication employ Extremely Large Multiple Input Multiple Output (ELMIMO) antenna arrays at the Base Station (BS) to expand the Radiative Near Field Region (RNFR) boundary by several meters. The existing Channel Estimation (CE) schemes in the Far Field Region (FFR) and RNFR had shown poor estimation accuracy at RNFR distance up to 1 km. In order to address this problem, RNFR boundary range within which a spherical wave can exists is deduced. A sparse three-dimensional (3D) Polar Depth Domain (PDD) codebook matrix has been developed using a new Depth Domain (DD) with finite range for each depth applicable within the obtained RNFR boundary range. A novel Compressive Sensing (CS) based CE algorithm named Depth Domain Orthogonal Matching Pursuit algorithm (DDOMP) is developed using generated 3D PDD codebook matrix. The Normalized Mean Square Error (NMSE) performance of the proposed algorithm has been analyzed in terms of depth, Pilot Length (PL), and Signal to Noise Ratio (SNR). It has been found that the DDOMP algorithm improves the CE accuracy when compared to the existing FFR and NFR CE schemes. It significantly reduces pilot overhead too. The performance of the DDOMP CE scheme is validated by simulation results.