Joint Array Diagnosis and Channel Estimation for RIS-Aided mmWave MIMO System

• Published in: IEEE access Vol. 8; pp. 193992 - 194006
• Main Authors: Li, Binrui, Zhang, Zhongpei, Hu, Zhenzhen, Chen, Yang
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 2020; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Antenna arrays; Array signal processing; Atmospheric effects; Beamforming; Channel estimation; Computational geometry; Convexity; Diagnosis; Estimation; joint array diagnosis and channel estimation; Millimeter waves; MIMO (control systems); MIMO communication; mmWave MIMO system; Noise reduction; Optimization; Parameter estimation; reconfigurable intelligent surface; Signal processing; Signal processing algorithms; Time; two-timescale non-convex optimization; Water drops; Weather
• ISSN: 2169-3536; 2169-3536
• DOI: 10.1109/ACCESS.2020.3032775

In this paper, we consider a reconfigurable intelligent surface (RIS) aided millimeter wave (mmWave) multiple-input multiple-output (MIMO) system. The system can obtain the huge gain via joint active beamforming at the base station (BS) and passive beamforming at the RIS. However, due to weather and atmospheric effects, outdoor RIS antenna elements are subject to full or partial blockages from a plethora of particles like dirt, salt, ice, and water droplets. These blockages can cause an approximate squared power/SNR loss for the system. Different from the conventional array diagnosis, the RIS has no signal processing capability. Thus, we propose the joint array diagnosis and channel estimation techniques containing two stages to solve the problem. At the first stage the channel parameters at user equipment (UE) and BS are estimated using an iterative reweighted (IR) method. At the second stage, the array blockage coefficient vector and the effective sparse channel parameters at RIS are jointly estimated via solving a two-timescale non-convex optimization problem. We propose two algorithms, i.e., a batch algorithm (BA) and a two-timescale online joint array diagnosis and channel estimation (TOJADCE) algorithm to solve the problem and compare the performance of these two algorithms. Finally, to speed up the convergence of long-term variable and improve estimation performance, we propose a noise reduction (NR) algorithm. The simulations verify the effectiveness of our proposed algorithms.