Multi-IRS-Aided Multi-User MIMO in Mmwave/thz Communications: A Space-Orthogonal Scheme

Boyu Ning,Peilan Wang,Lingxiang Li,Zhi Chen,Jun Fang

DOI: https://doi.org/10.1109/tcomm.2022.3216344

IF: 6.166

2022-01-01

IEEE Transactions on Communications

Abstract:Multiple-input multiple-output (MIMO) and intelligent reflecting surface (IRS) are two appealing technologies in millimeter-wave (mmWave) and terahertz (THz) communications. The challenge of combining these two technologies lies in joint design for active beamforming (at the base-station (BS)/users) and passive beamforming (at the IRSs). In this paper, we consider a multi-IRS-aided multi-user MIMO scenario and propose a novel space-orthogonal scheme by applying zero-forcing techniques. Specifically, we first propose a multi-IRS-based zero-interference criterion, under which multi-user interference can be eliminated regardless of the IRS’s phase shifts. Based on this criterion, we decompose the precoder/decoder matrix into a product of two matrices, with one of them devised for interference cancellation and the other one of them devised for achievable rate maximization. Next, an approximate space-orthogonal technique referred to as partial zero-forcing (IRS-PZF) is proposed for proposed for devising the former matrix whose objective is to cancel the multi-user interference; while two efficient phase-shift schemes are proposed for the IRS passive beamforming, namely, water-filling segment matching (WSM) and phase iterative evolution (PIE), which balance between performance and complexity. Finally, we calculate the latter matrix of the precoder/decoder by applying the singular value decomposition (SVD) for the effective BS-user channels, so as to maximize the users’ achievable rates. Numerical results demonstrate the effectiveness and superiority of our proposed scheme compared with the benchmarks.