QoS-Aware Hybrid Beamforming and DOA Estimation in Multi-Carrier Dual-Function Radar-Communication Systems

• Published in: IEEE journal on selected areas in communications Vol. 40; no. 6; pp. 1890 - 1905
• Main Authors: Cheng, Ziyang, Liao, Bin
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.06.2022; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Array signal processing; Beamforming; Communication; Communications systems; consensus-ADMM; Design; Direction of arrival; Direction-of-arrival estimation; DOA estimation; dual-function radar-communication (DFRC); Estimation; Hybrid beamforming (HBF); Interference; Optimization; Orthogonality; QoS constraint; Quality of service; Radar; Radio frequency; Signal classification; Waveforms
• ISSN: 0733-8716; 1558-0008
• DOI: 10.1109/JSAC.2022.3155529

In this paper, the issues of transmit hybrid beamforming (HBF) design and direction-of-arrival (DOA) estimation in multi-carrier dual-function radar-communication (DFRC) systems, in consideration of quality-of-service (QoS) for multiple users (MUs), are investigated. In the designed system, communication symbols are embedded into radar pulse interval with multiple orthogonal waveforms, and the HBF is optimized to focus the transmit energy within the spatial sectors of interest by taking the QoS requirement for MUs into account. The problem involving these considerations is formulated as the minimization of mean squared error (MSE) between the achieved spatial spectrum and a desired one, subject to constraints of communication QoS, constant modulus, power and orthogonality. Accordingly, a consensus alternating direction method of multipliers (consensus-ADMM) framework based on weighted mean-square error minimization (WMMSE) is devised to tackle the resultant nonconvex problem. Further, the closed-form solutions of the primal variables are derived. Additionally, the MUltiple SIgnal Classification (MUSIC)-based DOA estimation with the designed HBF architecture is presented, and the corresponding Cramér-Rao bound is derived. Numerical simulations are performed to demonstrate the effectiveness of the proposed designs.