Precoding and Detection for Broadband Single Carrier Terahertz Massive MIMO Systems Using LSQR Algorithm

• Published in: IEEE transactions on wireless communications Vol. 18; no. 2; pp. 1026 - 1040
• Main Authors: Peng, Bile, Wesemann, Stefan, Guan, Ke, Templ, Wolfgang, Kurner, Thomas
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.02.2019; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Antenna arrays; Antennas; Bandwidth; Bandwidths; Bit error rate; block matrix; Broadband; Broadband antennas; Broadband communication; Communication; Computer simulation; detection; Error detection; frequency selective channel; Frequency spectrum; Impulse response; Interference; least square QR (LSQR) algorithm; LSQR algorithm; massive MIMO; Millimeter waves; MIMO (control systems); MIMO communication; minimum mean square error; MMSE; Precoding; Propagation; Propagation losses; QR algorithms; Selectivity; Signal processing; Signal transmission; THz communication; Wireless communications
• ISSN: 1536-1276; 1558-2248; 1558-2248
• DOI: 10.1109/TWC.2018.2889056

The terahertz (THz) communication utilizes the frequency spectrum above 300 GHz and is widely considered as a promising solution to the future high-speed short-range wireless communication beyond millimeter wave communication. While providing tens of gigahertz bandwidth, it is subjected to high-propagation path loss, inter-symbol, and inter-user interferences. The massive multiple-input-multiple-output (MIMO) can be applied to address these problems by cooperation between many access point antennas. However, the THz channel characteristics, including high-propagation path loss, frequency selectivity, and a big number of samples per channel impulse response, require carefully tailored algorithm for massive MIMO signal processing. In this paper, we propose a single carrier minimum mean square error precoding and detection algorithm for frequency selective THz channels. The MIMO signal transmission is described with the block matrices. A gain control heuristic is introduced to reduce the complexity. The sparsity property of the channel is utilized to construct sparse channel matrices and the least square QR algorithm is applied to efficiently solve the problems. Besides the uniform antenna array, the hybrid array consisting of several subarrays is considered as well. The simulation results show that the massive MIMO array can provide a satisfactory performance in terms of bit error rate.