Low‐complexity linear precoding for low‐PAPR massive MU‐MIMO‐OFDM downlink systems

Abstract

To design new revolutionary wireless communication systems, orthogonal frequency-division multiplexing (OFDM)-based massive multiuser (MU) multiple-input multiple-output (MIMO) has been shown to be the most promising technology to significantly enhance the spectral, energy, and hardware efficiencies. However, massive MU-MIMO-OFDM transmitters exhibit signal with high peak-to-average power ratio (PAPR). Accordingly, the nonlinearity of the radio frequency (RF) power amplifier (PA), which causes the most severe hardware impairment, is expected to be a low-cost and energy-efficient component to enable cost-and energy-efficient massive MU-MIMO-OFDM BS deployments, generating harmful in-band distortion and out-of-band (OOB) emissions. In this paper, we develop a PAPR-aware downlink transmission scheme in an OFDM-based large-scale MU-MIMO. Linear precoding of data and peak-canceling signals (PCSs) are employed to reduce the PAPRs of the transmitted signals by exploiting the excess degrees of freedom (DoFs) provided by equipping the base station (BS) by a large number of transmit antennas. Specifically, we design PCSs to be added to the frequencydomain precoded data signals, with the goal of reducing the PAPRs of their time-domain counterpart signals. Most importantly, the added PCSs have to lie in the null spaces of their associated MIMO channel matrices such that they do not cause any MU interference (MUI) and OOB radiation. In this regard, an efficient algorithm is developed, which is based on different data and PCSs precoders, and the corresponding achievable PAPR reduction and bit error rate (BER) performance are analyzed. Moreover, to optimize a tradeoff between performance and complexity, linear precoders based on matrix polynomials (M-POLYs) and gradient-iterative approaches are studied for both data and PCSs precoding. Simulation results reveal that these latter provide similar performance as the regularized zero-forcing (RZF) and orthogonal projection null space (OPNS)-based data and PCSs precoders, while they need much lower computational complexity. The substantial PAPR reduction provided by the proposed algorithm offers interesting insights for the design of energy-efficient massive MU-MIMO-OFDM systems.

End-to-End Deep Learning for Multipair Two-Way Massive MIMO with PA Impairments

“…Recently, more interesting techniques have been proposed for massive MIMO systems, like joint MU precoding and PAPR reduction schemes introduced in [5] and [18], referred to as MU-PPGDm, RZF-OPNS, POLY-POLY-Horner, and POLY-OPNS. They consist in gradient-iterative method-based linear precoding.…”

“…8 presents the impact of the channel estimation error on the SER performance. We consider the channel matrix in the frequency domain as represented by: Ĥd = H d + Hd , where Hd ∼ CN (0, σ 2 h d ) is the channel estimation error whose entries are complex Gaussian distributed with a variance of σ 2 h d [18]. Figs.…”

Toward Energy-Efficient 6G Networks: Uplink Cell-Free Massive MIMO With NLD Cancellation Technique of Hardware Impairments

“…The key simulation parameters are summarized in Table 4. The parameters are chosen to align with those used in the literature [18]. The parameters of PAs are set as follows G = 16, Vsat = 1.9, p = 1.1, A = -345, B = 0.17 and q = 4.…”

Deep learning‐assisted reconfigurable intelligent surface for enhancing 6G mobile networks

“…For example, to efficiently use the MIMO antenna arrays in communication, one exclusive radio frequency band is needed with multiple active components such as the Analog to Digital Converter (ADC), frequency up‐converter, low‐noise amplifier, and per active antenna element, which consumes immense power and cost, specifically at mmWave spectrum. To tackle these problems in future mobile networks, a lot of researchers have discussed energy efficient and low complexity and cost techniques to sustain the high performance of the system, by utilizing simple hardware components with techniques that are better energy efficient in addition to the allocation of the green resource and design of transceiver signal processing methods 2–4 . In addition, using the massive MIMO architecture is a potential Fifth Generation (5G) mobile network technology by exploiting the extensive spatial multiplexing to provide the system with high spectral efficiency 5–9 …”

“…To tackle these problems in future mobile networks, a lot of researchers have discussed energy efficient and low complexity and cost techniques to sustain the high performance of the system, by utilizing simple hardware components with techniques that are better energy efficient in addition to the allocation of the green resource and design of transceiver signal processing methods. [2][3][4] In addition, using the massive MIMO architecture is a potential Fifth Generation (5G) mobile network technology by exploiting the extensive spatial multiplexing to provide the system with high spectral efficiency. [5][6][7][8][9] RIS is considered a great future mobile network key enabler for extending the coverage area by increasing the diversity of the paths that can overcome the effects of blockages in mmWave and Terahertz (THz) mobile networks.…”