Compression of Site-Specific Deep Neural Networks for Massive MIMO Precoding
Polytechnique Montréal（蒙特利尔理工学院）
Ghazal Kasalaee, Ali Hasanzadeh Karkan, Jean-François Frigon, François Leduc-Primeau
备注：
This preprint comprises 6 pages and features 3 figures. It has been accepted to the IEEE International Conference on Machine Learning and Computer Networking (ICMLCN) 2025
The deployment of deep learning (DL) models for precoding in massive multiple-input multiple-output (mMIMO) systems is often constrained by high computational demands and energy consumption. In this paper, we investigate the compute energy efficiency of mMIMO precoders using DL-based approaches, comparing them to conventional methods such as zero forcing and weighted minimum mean square error (WMMSE). Our energy consumption model accounts for both memory access and calculation energy within DL accelerators. We propose a framework that incorporates mixed-precision quantization-aware training and neural architecture search to reduce energy usage without compromising accuracy. Using a ray-tracing dataset covering various base station sites, we analyze how site-specific conditions affect the energy efficiency of compressed models. Our results show that deep neural network compression generates precoders with up to 35 times higher energy efficiency than WMMSE at equal performance, depending on the scenario and the desired rate. These results establish a foundation and a benchmark for the development of energy-efficient DL-based mMIMO precoders.
In this paper, we propose a novel joint caching and massive multiple-input multiple-output (MIMO) transmission scheme, referred to as cache-aided massive MIMO, for advanced downlink cellular communications. In addition to reaping the conventional advantages of caching and massive MIMO, the proposed scheme also exploits the side information provided by cached files for interference cancellation at the receivers. This interference cancellation increases the degrees of freedom available for precoding design. In addition, the power freed by the cache-enabled offloading can benefit the transmission…
Precoding for multibeam satellite systems with full frequency reuse in a multi-user multiple-input multiple-output (MU-MIMO) downlink scenario is addressed. A testbed is developed to perform an over-the-air field trial of zero forcing precoding for the spatial multiplexing of two separate video streams over two co-located geostationary satellites. Commercial-off-the-shelf DVB-S2x receivers in two single-antenna user terminals (UTs) are operated to successfully decode two independent video streams. To this end, particular attention is paid to a comprehensive assessment of the practical synchron…
Millimeter wave (mmWave) massive multiple-input multiple-output (MIMO) has been regarded to be an emerging solution for the next generation of communications, in which hybrid analog and digital precoding is an important method for reducing the hardware complexity and energy consumption associated with mixed signal components. However, the fundamental limitations of the existing hybrid precoding schemes is that they have high computational complexity and fail to fully exploit the spatial information. To overcome these limitations, this paper proposes, a deep-learning-enabled mmWave massive MIMO…
A novel precoding method based on supervised deep neural networks is introduced for the multiple-input multiple-output Gaussian wiretap channel. The proposed deep learning (DL)-based precoding learns the input covariance matrix through offline training over a large set of input channels and their corresponding covariance matrices for efficient, reliable, and secure transmission of information. Furthermore, by spending time in offline training, this method remarkably reduces the computation complexity in real-time applications. Compared to traditional precoding methods, the proposed DL-based pr…
In this work we investigate ultra-reliable low-latency massive multiple-input multiple-output (MIMO) communication links in vehicular scenarios, where coherence between uplink and downlink cannot be assumed. In such scenarios the channel state information obtained in the uplink will be outdated for the following downlink phase. To compensate for this channel aging we will utilize orthogonal precoding with two-dimensional precoding sequences in the time-frequency domain within an orthogonal frequency division multiplexing system. The channel hardening effect of massive MIMO transmission decreas…
Coarsely quantized MIMO signalling methods have gained popularity in the recent developments of massive MIMO as they open up opportunities for massive MIMO implementation using cheap and power-efficient radio-frequency front-ends. This paper presents a new one-bit MIMO precoding approach using spatial Sigma-Delta ($ΣΔ$) modulation. In previous one-bit MIMO precoding research, one mainly focuses on using optimization to tackle the difficult binary signal optimization problem that arises from the precoding design. Our approach attempts a different route. Assuming angular MIMO channels, we apply …
Precoding is a method of compensating the channel at the transmitter. This work presents a novel method of data detection in turbo coded single user massive multiple input multiple output (MIMO) systems using precoding. We show via computer simulations that, when precoding is used, re-transmitting the data does not result in significant reduction in bit-error-rate (BER), thus increasing the spectral efficiency, compared to the case without precoding. Moreover, increasing the number of transmit and receive antennas results in improved BER.
Using precoding to suppress multi-user interference is a well-known technique to improve spectra efficiency in multiuser multiple-input multiple-output (MU-MIMO) systems, and the pursuit of high performance and low complexity precoding method has been the focus in the last decade. The traditional algorithms including the zero-forcing (ZF) algorithm and the weighted minimum mean square error (WMMSE) algorithm failed to achieve a satisfactory trade-off between complexity and performance. In this paper, leveraging on the power of deep learning, we propose a low-complexity precoding design framewo…
Considering the expensive radio frequency (RF) chain, huge training overhead and feedback burden issues in massive MIMO, in this letter, we propose a mixed-timescale per-group hybrid precoding (MPHP) scheme under an adaptive partially-connected RF precoding structure (PRPS), where the RF precoder is implemented using an adaptive connection network (ACN) and M analog phase shifters (APSs), where M is the number of antennas at the base station (BS). Exploiting the mixed-time stage channel state information (CSI) structure, the joint-design of ACN and APSs is formulated as a statistical signal-to…
Signal processing in wireless communications, such as precoding, detection, and channel estimation, are basically about solving inverse matrix problems, which, however, are slow and inefficient in conventional digital computers, thus requiring a radical paradigm shift to achieve fast, real-time solutions. Here, for the first time, we apply the emerging analog matrix computing (AMC) to the linear precoding of massive MIMO. The real-valued AMC concept is extended to process complex-valued signals. In order to adapt the MIMO channel models to RRAM conductance mapping, a new matrix inversion circu…
In this letter, we consider linear precoding for downlink massive multi-user (MU) multiple-input multiple-output (MIMO) systems. We propose the novel successively-regularized zero forcing (SRZF) precoding, which exploits successive null spaces of the MIMO channels of the users, along with regularization, to control the inter-user interference (IUI) and to enhance performance and robustness to imperfect channel state information (CSI) at the base station (BS). We study the IUI characteristics of the proposed SRZF precoding for perfect and imperfect CSI at the BS. Furthermore, via computer simul…
This paper investigates the hybrid precoding design for millimeter wave (mmWave) multiple-input multiple-output (MIMO) systems with finite-alphabet inputs. The precoding problem is a joint optimization of analog and digital precoders, and we treat it as a matrix factorization problem with power and constant modulus constraints. Our work presents three main contributions: First, we present a sufficient condition and a necessary condition for hybrid precoding schemes to realize unconstrained optimal precoders exactly when the number of data streams Ns satisfies Ns = minfrank(H);Nrfg, where H rep…
Cell densification is a key driver to increase area spectral efficiencies in multi-antenna cellular systems. As increasing the densities of base stations (BSs) and users that share the same spectrum, however, both inter-user-interference (IUI) and inter-cell interference (ICI) problems give rise to a significant loss in spectral efficiencies in such systems. To resolve this problem under the constraint of local channel state information per BS, in this paper, we present a novel noncooperative multi-user multiple-input multiple-output (MIMO) precoding technique, called signal-to-interference-pu…
In this paper we study on a massive MIMO relay system with linear precoding under the conditions of imperfect channel state information at the transmitter (CSIT) and per-user channel transmit correlation. In our system the source-relay channels are massive multiple-input multiple-output (MIMO) ones and the relay-destination channels are massive multiple-input single-output (MISO) ones. Large random matrix theory (RMT) is used to derive a deterministic equivalent of the signal-to-interference-plus-noise ratio (SINR) at each user in massive MIMO amplify-forward and decode-forward (M-MIMO-ADF) re…
In both academia and industry, multi-user multiple-input multiple-output (MU-MIMO) techniques have shown enormous gains in spectral efficiency by exploiting spatial degrees of freedom. So far, an underlying assumption in most of the existing MU-MIMO design has been that all the users use infinite blocklength, so that they can achieve the Shannon capacity. This setup, however, is not suitable considering delay-constrained users whose blocklength tends to be finite. In this paper, we consider a heterogeneous setting in MU-MIMO systems where delay-constrained users and delay-tolerant users coexis…
We present a novel and low-complexity massive multiple-input multiple-output (MIMO) precoding strategy based on novel findings concerning the subspace separability of Rician fading channels. Considering a uniform rectangular array at the base station, we show that the subspaces spanned by the channel vectors can be factorized as a tensor product between two lower dimensional subspaces. Based on this result, we formulate tensor maximum ratio transmit and zero-forcing precoders. We show that the proposed tensor precoders exhibit lower computational complexity and require less instantaneous chann…
Constant envelope (CE) precoding design is of great interest for massive multiuser multi-input multi-output systems because it can significantly reduce hardware cost and power consumption. However, existing CE precoding algorithms are hindered by excessive computational overhead. In this letter, a novel model-driven deep learning (DL)-based network that combines DL with conjugate gradient algorithm is proposed for CE precoding. Specifically, the original iterative algorithm is unfolded and parameterized by trainable variables. With the proposed architecture, the variables can be learned effici…
One of the first widespread uses of multi-user multiple-input multiple-output (MU-MIMO) is in 5G networks, where each base station has an advanced antenna system (AAS) that is connected to the baseband unit (BBU) with a capacity-constrained fronthaul. In the AAS configuration, multiple passive antenna elements and radio units are integrated into a single box. This paper considers precoded downlink transmission over a single-cell MU-MIMO system. We study optimized linear precoding for AAS with a limited-capacity fronthaul, which requires the precoding matrix to be quantized. We propose a new pr…
In this letter, we study the performance of a downlink multiuser massive multiple-input multiple-output (MIMO) system with sub-connected structure over limited feedback channels. Tight rate approximations are theoretically analyzed for the system with pure analog precoding and hybrid precoding. The effect of quantized analog and digital precoding is characterized in the derived expressions. Furthermore, it is revealed that the pure analog precoding outperforms the hybrid precoding using maximal-ratio transmission (MRT) or zero forcing (ZF) under certain conditions, and we theoretically charact…
This paper investigates the use of a reconfigurable intelligent surface (RIS) to aid point-to-point multi-data-stream multiple-input multiple-output (MIMO) wireless communications. With practical finite alphabet input, the reflecting elements at the RIS and the precoder at the transmitter are alternatively optimized to minimize the symbol error rate (MSER). In the reflecting optimization with a fixed precoder, two reflecting design methods are developed, referred as eMSER-Reflecting and vMSER-Reflecting. In the optimization of the precoding matrix with a fixed reflecting pattern, the matrix op…
Deploying antenna arrays with an asymptotically large aperture will be central to achieving the theoretical gains of massive MIMO in beyond-5G systems. Such extra-large MIMO (XL-MIMO) systems experience propagation conditions which are not typically observed in conventional massive MIMO systems, such as spatial non-stationarities and near-field propagation. Moreover, standard precoding schemes, such as zero-forcing (ZF), may not apply to XL-MIMO transmissions due to the prohibitive complexity associated with such a large-scale scenario. We propose two novel precoding schemes that aim at reduci…
Cell-free massive MIMO (CF-M-MIMO) systems represent an evolution of the classical cellular architecture that has dominated the mobile landscape for decades. In CF-M-MIMO, a central processing unit (CPU) controls a multitude of access points (APs) that are irregularly scattered throughout the coverage area effectively becoming a fully distributed implementation of the M-MIMO technology. As such, it inherits many of the key properties that have made M-MIMO one of the physical layer pillars of 5G systems while opening the door to new features not available in M-MIMO. Among the latest is the poss…
Massive multi-user multiple-input multiple-output (MU-MIMO) wireless systems operating at millimeter-wave (mmWave) frequencies enable simultaneous wideband data transmission to a large number of users. In order to reduce the complexity of MU precoding in all-digital basestation architectures, we propose a two-stage precoding architecture that first performs precoding using a sparse matrix in the beamspace domain, followed by an inverse fast Fourier transform that converts the result to the antenna domain. The sparse precoding matrix requires a small number of multipliers and enables regular ha…
We consider the vector perturbation (VP) precoder design for multiuser multiple-input single output (MU-MISO) broadcast channel systems which is robust to power scaling factor errors. VP precoding has so far been developed and analyzed under the assumption that receivers could have known the power scaling factor in advance of tranmission perfectly, which is hard to obtain due to the large dynamic range and limited feedforward. However, as demonstrated in our results the performance of VP precoding is quite sensitive to the accuracy of power scaling factor and always encounter an error floor at…
In this paper, a novel nonlinear precoding (NLP) technique, namely constellation-oriented perturbation (COP), is proposed to tackle the scalability problem inherent in conventional NLP techniques. The basic concept of COP is to apply vector perturbation (VP) in the constellation domain instead of symbol domain; as often used in conventional techniques. By this means, the computational complexity of COP is made independent to the size of multi-antenna (i.e., MIMO) networks. Instead, it is related to the size of symbol constellation. Through widely linear transform, it is shown that COP has its …
Massive multi-input multiple-out (MIMO) is a key ingredient in improving the spectral efficiencies for next-generation cellular systems. Thanks to the channel reciprocity, in time-division-duplexing mode, each base station (BS) can acquire local channel state information at the transmitter (CSIT) for a set of users possibly located in adjacent cells. When the small cell BSs equipped with not-so-many antennas are densely deployed with marcrocells, a simple noncooperative MIMO precoding technique using local CSIT fails to achieve high spectral efficiency because of strong inter-cell-interference…
The use of low-resolution digital-to-analog and analog-to-digital converters (DACs and ADCs) significantly benefits energy efficiency (EE) at the cost of high quantization noise in implementing massive multiple-input multiple-output (MIMO) systems. For maximizing EE in quantized downlink massive MIMO systems, this paper formulates a precoding optimization problem with antenna selection; yet acquiring the optimal joint precoding and antenna selection solution is challenging due to the intricate EE characterization. To resolve this challenge, we decompose the problem into precoding direction and…
Terahertz (THz) band communication has been widely studied to meet the future demand for ultra-high capacity. In addition, multi-input multi-output (MIMO) technique and non-orthogonal multiple access (NOMA) technique with multi-antenna also enable the network to carry more users and provide multiplexing gain. In this paper, we study the maximization of energy efficiency (EE) problem in THz-NOMA-MIMO systems for the first time. And the original optimization problem is divided into user clustering, hybrid precoding and power optimization. Based on channel correlation characteristics, a fast conv…
In this paper, we consider the precoder design for an underloaded or critically loaded downlink multi-user multiple-input multiple-output (MIMO) communication system. We propose novel precoding and decoding schemes which enhance system performance based on rate splitting at the transmitter and single-stage successive interference cancellation at the receivers. The proposed successive null-space (SNS) precoding utilizes linear combinations of the null-space basis vectors of the successively augmented MIMO channel matrices of the users as precoding vectors to adjust the inter-user-interference e…
Duzce University（杜兹切大学）；University of Hertfordshire（赫特福德大学）；University of Luxembourg（卢森堡大学）
In multi-user millimeter wave (mmWave) multiple-input-multiple-output (MIMO) systems, hybrid precoding is a crucial task to lower the complexity and cost while achieving a sufficient sum-rate. Previous works on hybrid precoding were usually based on optimization or greedy approaches. These methods either provide higher complexity or have sub-optimum performance. Moreover, the performance of these methods mostly relies on the quality of the channel data. In this work, we propose a deep learning (DL) framework to improve the performance and provide less computation time as compared to convention…
Channel estimation and hybrid precoding are considered for multi-user millimeter wave massive multi-input multi-output system. A deep learning compressed sensing (DLCS) channel estimation scheme is proposed. The channel estimation neural network for the DLCS scheme is trained offline using simulated environments to predict the beamspace channel amplitude. Then the channel is reconstructed based on the obtained indices of dominant beamspace channel entries. A deep learning quantized phase (DLQP) hybrid precoder design method is developed after channel estimation. The training hybrid precoding n…
Power consumption of multi-user (MU) precoding is a major concern in all-digital massive MU multiple-input multiple-output (MIMO) base-stations with hundreds of antenna elements operating at millimeter-wave (mmWave) frequencies. We propose to replace part of the linear Wiener filter (WF) precoding matrix by a finite-alphabet WF precoding (FAWP) matrix, which enables the use of low-precision hardware that consumes low power and area. To minimize the performance loss of our approach, we present methods that efficiently compute FAWP matrices that best mimic the WF precoder. Our results show that …
Hybrid precoding has been recognized as a promising technology to combat the path loss of millimeter wave signals in massive multiple-input multiple-output (MIMO) systems. However, due to the joint optimization of the digital and analog precoding matrices as well as extra constraints for the analog part, the hybrid precoding design is still a tough issue in current research. In this paper, we adopt the thought of clustering in unsupervised learning and provide design schemes for fully-connected hybrid precoding (FHP) and adaptively-connected hybrid precoding (AHP) in multi-user massive MIMO sy…
Massive multi-user multiple-input multiple-output (MU-MIMO) enables significant gains in spectral efficiency and link reliability compared to conventional small-scale MIMO technology. Furthermore, linear precoders, e.g., using zero forcing or Wiener filter (WF) precoding, are sufficient to achieve excellent error-rate performance in the massive MU-MIMO downlink. However, these methods necessitate centralized processing at the base-station (BS), which causes (i) excessively high interconnect and chip input/output data rates, and (ii) high implementation complexity. We propose two feedforward ar…
Deep Learning-Based Tri-Hybrid Multi-User MIMO Precoding: The Blessing of EM-Reconfigurable Antennas
Beijing Institute of Technology（北京理工大学）；BIT (Zhuhai)（北京理工大学（珠海））；Tsinghua University（清华大学）；Guilin University of Electronic Technology（桂林电子科技大学）；Yangtze Delta Region Academy, BIT (Jiaxing)（北京理工大学（嘉兴）长三角研究院）；Advanced Research Institute of Multidisciplinary Sciences, BIT（北京理工大学前沿交叉科学研究院）
Most works on cell-free massive multiple-input multiple-output (MIMO) consider non-cooperative precoding strategies at the base stations (BSs) to avoid extensive channel state information (CSI) exchange via backhaul signaling. However, considerable performance gains can be accomplished by allowing coordination among the BSs. This paper proposes the first distributed framework for cooperative precoding design in cell-free massive MIMO (and, more generally, in joint transmission coordinated multi-point) systems that entirely eliminates the need for backhaul signaling for CSI exchange. A novel ov…
Distributed phased arrays based multiple-input multiple-output (DPA-MIMO) is a recently proposed highly reconfigurable architecture enabling both spatial multiplexing and beamforming in millimeter-wave (mmWave) systems. In this work, we focus on coping with the hybrid precoding for the wideband DPA-MIMO system with orthogonal frequency division multiplexing (OFDM) modulation. More specifically, we propose an alternating direction method of multipliers (ADMM) enabled hybrid precoding approach based on an alternating optimization framework, abbreviated to ADMM-AltMin, for such cooperative array-…
Massive MIMO requires a large number of antennas and the same amount of power amplifiers (PAs), one per antenna. As opposed to 4G base stations, which could afford highly linear PAs, next-generation base stations will need to use inexpensive PAs, which have a limited region of linear amplification. One of the research challenges is effectively handling signals which have high peak-to-average power ratios (PAPRs), such as orthogonal frequency division multiplexing (OFDM). This paper introduces a PAPR-aware precoding scheme that exploits the excessive spatial degrees-of-freedom of large scale mu…
The high energy consumption of massive multi-input multi-out (MIMO) system has become a prominent problem in the millimeter wave(mm-Wave) communication scenario. The hybrid precoding technology greatly reduces the number of radio frequency(RF) chains by handing over part of the coding work to the phase shifting network, which can effectively improve energy efficiency. However, conventional hybrid precoding algorithms based on mathematical means often suffer from performance loss and high computational complexity. In this paper, a novel BP-neural-network-enabled hybrid precoding algorithm is pr…
Precoding design exploiting deep learning methods has been widely studied for multiuser multiple-input multiple-output (MU-MIMO) systems. However, conventional neural precoding design applies black-box-based neural networks which are less interpretable. In this paper, we propose a deep learning-based precoding method based on an interpretable design of a neural precoding network, namely iPNet. In particular, the iPNet mimics the classic minimum mean-squared error (MMSE) precoding and approximates the matrix inversion in the design of the neural network architecture. Specifically, the proposed …
Aiming at maximizing the achievable sum-rate of wideband multiuser mmWave massive MIMO systems, the hybrid precoding is studied. Since each computation of the achievable sum-rate can be performed only after the analog precoder and digital precoder are both determined, the maximization of the achievable sum-rate has intractable computational complexity. By introducing the interference free (IF) achievable sum-rate, the design of the analog and digital precoders can be decoupled. To avoid the beam conflict and maximize the IF achievable sum-rate, a Hungarian-based codeword selection algorithm is…
Massive multiuser multiple-input multiple-output (MU-MIMO) has been the mainstream technology in fifth-generation wireless systems. To reduce high hardware costs and power consumption in massive MU-MIMO, low-resolution digital-to-analog converters (DAC) for each antenna and radio frequency (RF) chain in downlink transmission is used, which brings challenges for precoding design. To circumvent these obstacles, we develop a model-driven deep learning (DL) network for massive MU-MIMO with finite-alphabet precoding in this article. The architecture of the network is specially designed by unfolding…
Hybrid precoding is a cost-efficient technique for millimeter wave (mmWave) massive multiple-input multiple-output (MIMO) communications. This paper proposes a deep learning approach by using a distributed neural network for hybrid analog-and-digital precoding design with limited feedback. The proposed distributed neural precoding network, called DNet, is committed to achieving two objectives. First, the DNet realizes channel state information (CSI) compression with a distributed architecture of neural networks, which enables practical deployment on multiple users. Specifically, this neural ne…
This study proposes a symbol-level precoding algorithm based on the minimum mean squared error design objective under a strict per antenna power constraint for PSK modulation. The proposed design is then formulated in the standard form of a second-order cone program, allowing for an optimal solution via the interior point method. Numerical results indicate that the proposed design is superior to the existing approaches in terms of bit-error-rate for the low and intermediate SNR regime.
