Channel State Information Compression Research Articles

IALNet: An Integration Attention Lightweight Neural Network for Massive MIMO CSI Feedback

Recent deep learning-based channel state information (CSI) feedback methods have made great progress. However, numerous existing methods improve CSI compression and reconstruction accuracy at the cost of computational complexity by designing more complex deep learning modules. In this letter, we propose a novel lightweight neural network IALNet for CSI feedback problem. In the proposed IALNet, we design an integration attention module (IAM) for improving the performance of the network. Specifically, By embedding the correlation information of the vertical and horizontal directions of CSI matrix into the channel attention, the IAM enabled IALNet to capture distribution characteristics of CSI matrix while focusing on the important regions of interest in the vertical and horizontal directions and enhancing the feature representation. Extensive experiment results demonstrate that our IALNet outperforms previous SOTA networks in both outdoor and indoor scenarios, providing an efficient, low cost and flexible CSI feedback solution.

Viewing Channel as Sequence Rather Than Image: A 2-D Seq2Seq Approach for Efficient MIMO-OFDM CSI Feedback

In this paper, we aim to design an effective learning-based channel state information (CSI) feedback scheme for the multiple-input multiple-output (MIMO) orthogonal frequency division multiplexing (OFDM) systems from a physics-inspired perspective. We first argue that the CSI matrix of a MIMO-OFDM system is physically closer to a two-dimensional (2-D) sequence rather than an image due to its apparent unsmoothness, non-scalability, and translational variance within both the spatial and frequency domains. On this basis, we introduce a 2-D long short-term memory (LSTM) neural network to represent the CSI and propose a 2-D sequence-to-sequence (Seq2Seq) model for CSI compression and reconstruction. Specifically, one two-layer 2-D LSTM is used for CSI feature extraction, and the other is used for CSI representation and reconstruction. The proposed scheme can not only fully utilize the unique 2-D characteristics of CSI but also preserve the index information and unsmooth features of the CSI matrix compared with current convolutional neural network (CNN) based schemes. We show that the computational complexity of the proposed scheme is linear in the number of transmit antennas and subcarriers. Its key performances, like reconstruction accuracy, convergence speed, generalization ability after short-term training, and robustness to lossy feedback, are comprehensively compared with existing popular convolutional networks. Experimental results show that our scheme can bring up to nearly 7 dB gain in reconstruction accuracy under the same overhead and reduce feedback overhead by up to 75% under the same accuracy compared with the conventional CNN-based approaches.

Self-Information Domain-Based Neural CSI Compression With Feature Coupling

Deep learning (DL)-based channel state information (CSI) feedback methods compressed the CSI matrix by exploiting its delay and angle features straightforwardly, while the measure in terms of information contained in the CSI matrix has rarely been considered. Based on this observation, we introduce self-information as an informative CSI representation from the perspective of information theory, which reflects the amount of information of the original CSI matrix in an explicit way. Then, a novel DL-based network is proposed for temporal CSI compression in the self-information domain, namely SD-CsiNet. The proposed SD-CsiNet projects the raw CSI onto a self-information matrix in the newly-defined self-information domain, extracts both temporal and spatial features of the self-information matrix, and then couples these two features for effective compression. Experimental results verify the effectiveness of the proposed SD-CsiNet by exploiting the self-information of CSI. Particularly for compression ratios <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$1/8$</tex-math></inline-formula> and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$1/16$</tex-math></inline-formula> , the SD-CsiNet respectively achieves 7.17 dB and 3.68 dB performance gains compared to state-of-the-art methods.

Dual-Propagation-Feature Fusion Enhanced Neural CSI Compression for Massive MIMO

Due to the ability of feature extraction, deep learning (DL)-based methods have been recently applied to channel state information (CSI) compression feedback in massive multiple-input multiple-output (MIMO) systems. Existing DL-based CSI compression methods are usually effective in extracting a certain type of features in the CSI. However, the CSI usually contains two types of propagation features, i.g., non-line-of-sight (NLOS) propagation-path feature and dominant propagation-path feature, especially in channel environments with rich scatterers. To fully extract the both propagation features and learn a dual-feature representation for CSI, this paper proposes a dual-feature-fusion neural network (NN), referred to as DuffinNet. The proposed DuffinNet adopts a parallel structure with a convolutional neural network (CNN) and an attention-empowered neural network (ANN) to respectively extract different features in the CSI, and then explores their interplay by a fusion NN. Built upon this proposed DuffinNet, a new encoder-decoder framework is developed, referred to as Duffin-CsiNet, for improving the end-to-end performance of CSI compression and reconstruction. To facilitate the application of Duffin-CsiNet in practice, this paper also presents a two-stage approach for codeword quantization of the CSI feedback. Besides, a transfer learning-based strategy is introduced to improve the generalization of Duffin-CsiNet, which enables the network to be applied to new propagation environments. Simulation results illustrate that the proposed Duffin-CsiNet noticeably outperforms the existing DL-based methods in terms of reconstruction performance, encoder complexity, and network convergence, validating the effectiveness of the proposed dual-feature fusion design.

CSI Compression for Distributed-MIMO With Centralized Precoding and Power Allocation

In distributed multiple-input multiple-output (D-MIMO), a large number of distributed antenna points (APs) are coordinated by a central processing unit (CPU) to serve a limited number of users at the same time/frequency resources, which brings improvements in spectral efficiency. The success of D-MIMO relies on precoding and power allocation, where they can be performed fully centralized on the CPU or distributed in the APs. The centralized approach presents higher spectral efficiency than the distributed implementation but requires a significant peak fronthaul traffic due to channel state information (CSI) exchange between APs and CPU. In this paper, we propose CSI compression schemes in order to enable practical centralized D-MIMO implementation. It is shown that depending on the compression configuration, the spectral efficiency can be as good as the case without compression.

A Spatially Separable Attention Mechanism for Massive MIMO CSI Feedback

Channel State Information (CSI) Feedback plays a crucial role in achieving higher gains through beamforming. However, for a massive MIMO system, this feedback overhead is huge and grows linearly with the number of antennas. To reduce the feedback overhead several compressive sensing (CS) techniques were implemented in recent years but these techniques are often iterative and are computationally complex to realize in power-constrained user equipment (UE). Hence, a data-based deep learning approach took over in these recent years introducing a variety of neural networks for CSI compression. Specifically, transformer-based networks have been shown to achieve state-of-the-art performance. However, the multi-head attention operation, which is at the core of transformers, is computationally complex making transformers difficult to implement on a UE. In this work, we present a lightweight transformer named STNet which uses a spatially separable attention mechanism that is significantly less complex than the traditional full-attention. Equipped with this, STNet outperformed state-of-the-art models in some scenarios with approximately $1/10^{th}$ of the resources.

Deep Joint Source-Channel Coding for CSI Feedback: An End-to-End Approach

The increased throughput brought by MIMO technology relies on the knowledge of channel state information (CSI) acquired in the base station (BS). To make the CSI feedback overhead affordable for the evolution of MIMO technology (e.g., massive MIMO and ultra-massive MIMO), deep learning (DL) is introduced to deal with the CSI compression task. In traditional communication systems, the compressed CSI bits is treated equally and expected to be transmitted accurately over the noisy channel. While the errors occur due to the limited bandwidth or low signal-to-noise ratios (SNRs), the reconstruction performance of the CSI degrades drastically. As a branch of semantic communications, deep joint source-channel coding (DJSCC) scheme performs better than the separate source-channel coding (SSCC) scheme—the cornerstone of traditional communication systems—in the limited bandwidth and low SNRs. In this paper, we propose a DJSCC based framework for the CSI feedback task. In particular, the proposed method can simultaneously learn from the CSI source and the wireless channel. Instead of truncating CSI via Fourier transform in the delay domain in existing methods, we apply non-linear transform networks to compress the CSI. Furthermore, we adopt an SNR adaption mechanism to deal with wireless channel variations. The extensive experiments demonstrate the validity, adaptability, and generality of the proposed framework.

Overview of Deep Learning-Based CSI Feedback in Massive MIMO Systems

Many performance gains achieved by massive multiple-input and multiple-output depend on the accuracy of the downlink channel state information (CSI) at the transmitter (base station), which is usually obtained by estimating at the receiver (user equipment) and feeding back to the transmitter. The overhead of CSI feedback occupies substantial uplink bandwidth resources, especially when the number of transmit antennas is large. Deep learning (DL)-based CSI feedback refers to CSI compression and reconstruction by a DL-based autoencoder and can greatly reduce feedback overhead. In this paper, a comprehensive overview of state-of-the-art research on this topic is provided, beginning with basic DL concepts widely used in CSI feedback and then categorizing and describing some existing DL-based feedback works. The focus is on novel neural network architectures and utilization of communication expert knowledge to improve CSI feedback accuracy. Works on joint design of CSI feedback with other communication modules are also introduced, and some practical issues, including bitstream generation, multirate feedback, imperfect feedback, NN complexity, training dataset collection, online training, and standardization effect, are discussed. At the end of the paper, some challenges and potential research directions associated with DL-based CSI feedback in future wireless communication systems are identified.

Quan-Transformer Based Channel Feedback for RIS-Aided Wireless Communication Systems

Reconfigurable intelligent surface (RIS) has become one of the key technologies for future 6G communications due to its characteristics of an intelligent and controllable wireless environment. In the frequency division duplex mode, the user equipment feeds back the downlink channel state information (CSI) to the base station through the feedback link, to obtain potential gains for the RIS-aided wireless communication system. A large number of RIS reflection elements greatly increase the feedback link cost of the system. In this letter, Quan-Transformer, a sample network framework for CSI compression and reconstruction, is proposed based on Transformer network. The proposed framework can not only compress CSI effectively but also recover CSI with high precision, which improves the performance of the RIS-aided wireless communication system. According to the experimental results, compared with the CsiNet scheme, the achievable sum-rate and normalized mean squared error of the proposed scheme are increased by 6.57% and 3.32%, respectively.

Self-Supervised Learning for CSI Compression in FDD Massive MIMO Systems

In frequency division duplexing systems, the user equipment frequently reports the channel state information (CSI) to the base station. This information is crucial for link adaptation and beamforming tasks in massive MIMO systems. However, it consumes a large amount of bandwidth when the number of antennas and subcarriers increases. Currently, compression techniques have been used to reduce this bandwidth overhead. In this letter, we propose a novel learning-based technique for CSI compression. Our method exploits the bias/variance tradeoff for CSI compression based on a shallow neural network to approximate a sufficient statistic function for each sampled channel. Unlike prior work that adopts projection-based techniques for compression, our method interprets the model weights as the compressed representation. The experimental results confirm the advantages of the proposed method compared with other state-of-the-art models.

Deep CSI Compression for Massive MIMO: A Self-Information Model-Driven Neural Network

In order to fully exploit the advantages of massive multiple-input multiple-output (mMIMO), it is critical for the transmitter to accurately acquire the channel state information (CSI). Deep learning (DL)-based methods have been proposed for CSI compression and feedback to the transmitter. Although most existing DL-based methods consider the CSI matrix as an image, structural features of the CSI image are rarely exploited in neural network design. As such, we propose a model of self-information that dynamically measures the amount of information contained in each patch of a CSI image from the perspective of structural features. Then, by applying the self-information model, we propose a model-and-data-driven network for CSI compression and feedback, namely IdasNet. The IdasNet includes the design of a module of self-information deletion and selection (IDAS), an encoder of informative feature compression (IFC), and a decoder of informative feature recovery (IFR). In particular, the model-driven module of IDAS pre-compresses the CSI image by removing informative redundancy in terms of the self-information. The encoder of IFC then conducts feature compression to the pre-compressed CSI image and generates a feature codeword which contains two components, i.e., codeword values and position indices of the codeword values. Subsequently, the IFR decoder decouples the codeword values as well as position indices to recover the CSI image. Experimental results verify that the proposed IdasNet noticeably outperforms existing DL-based networks under various compression ratios while it has the number of network parameters reduced by orders-of-magnitude compared with various existing methods.

Distributed Neural Precoding for Hybrid mmWave MIMO Communications With Limited Feedback

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 <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">DNet</i> , 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 network is composed of multiple independent sub-networks with the same structure and parameters, which reduces both the number of training parameters and network complexity. Secondly, DNet learns the calculation of hybrid precoding from reconstructed CSI from limited feedback. Different from existing black-box neural network design, the DNet is specifically designed according to the data form of the matrix calculation of hybrid precoding. Simulation results show that the proposed DNet significantly improves the performance up to nearly 50% compared to traditional limited feedback precoding methods under the tests with various CSI compression ratios.

A Novel Approach Using Convolutional Transformer for Massive MIMO CSI Feedback

In frequency division duplex model, the massive multiple-input multiple-output (MIMO) systems rely on feedback of channel state information (CSI) to perform precoding operations. This can increase the potential transmission gain of the system. The use of large scale antennas increases the feedback overhead of CSI exponentially. Therefore, CSI must be compressed for feedback. However, convolution based feedback methods lack long-range dependency modeling of the CSI, resulting in limited effects at high compression ratios. In this letter, we propose a novel neural network based on the convolutional transformer architecture to improve the performance of CSI compression and reconstruction at high compression rates. The experimental results show that the average accuracy of ours is improved by 5.39&#x0025; over the state-of-the-art method at a high compression rate of 1/64. At the same time, the overall performance of the system has been improved.

ML-Based Massive MIMO Channel Prediction: Does It Work on Real-World Data?

Accurate channel state information (CSI) acquisition is hindered by CSI estimation errors, compression, feedback, and processing delays. We propose a machine learning (ML)-based massive multiple-input multiple-output (mMIMO) channel predictor (CP), which can work on the estimated channel and the compressed version of the estimated channel as well. While existing work has evaluated the performance of ML algorithms by only using the artificially generated channel realizations, this letter reports the results of the ML algorithm using the real-world channel realizations from a measurement campaign performed at Nokia Bell-Labs. The results corroborate the validity of the proposed ML-based CP.

LCF: A Deep Learning-Based Lightweight CSI Feedback Scheme for MIMO Networks

Recently, as deep learning technologies have received much attention for their great potential in extracting the principal components of data, there have been many efforts to apply them to the Channel State Information (CSI) feedback overhead problem, which can significantly limit Multi-Input Multi-Output (MIMO) beamforming gains. Unfortunately, since most compression models can quickly become outdated due to channel variation, timely model updates are essential for reflecting the current channel conditions, resulting in frequent additional transmissions for model sharing between transceivers. In particular, the heavy network models employed by most previous studies to achieve high compression gains exacerbate the impact of the overhead, eventually cancelling out the benefits of deep learning-based CSI compression. To address these issues, in this paper, we propose Lightweight CSI Feedback (LCF), a new lightweight CSI feedback scheme. LCF fully utilizes autoregressive Long Short-Term Memory (LSTM) to generate CSI predictions and uses them to train the autoencoder, so that the compression model could work effectively even in highly dynamic wireless channels. In addition, 3D convolutional layers are directly adopted in the autoencoder to capture diverse types of channel correlations in three dimensions. Extensive experiments show that LCF achieves a lower CSI compression error in terms of the Mean Squared Error (MSE), using only about 10% of the overhead of existing approaches.

ACCsiNet: Asymmetric Convolution-Based Autoencoder Framework for Massive MIMO CSI Feedback

Channel state information (CSI) is a critical part for massive multiple-input multiple-output (MIMO) system. However, it is a big challenge to send a large amount of CSI from the receiver to the transmitter with limited channel resources. In this letter, we propose asymmetric convolution-based autoencoder framework (ACCsiNet) to handle the CSI compression and decompression problem. Specifically, asymmetric convolution block (AC-Block) is used to enhance the feature extraction ability of convolution. Further, a lightweight method is applied, which can greatly reduce the storage space at the receiver. Considering the practical deployment, multi-model fusion schemes including multi-rate and multi-scenario fusion are also discussed to strengthen the generalization ability of the network. Experimental results show that the proposed ACCsiNet can improve the NMSE and cosine similarity <inline-formula> <tex-math notation="LaTeX">$\rho $ </tex-math></inline-formula> performance, especially for outdoor scenario. The results also verify that both the lightweight and multi-model fusion schemes can reach a near-optimal performance of the proposed ACCsiNet, but further significantly reduce the parameter amount by more than 83&#x0025; and 90&#x0025;, respectively.

Deep Learning Based CSI Compression and Quantization With High Compression Ratios in FDD Massive MIMO Systems

Generally, the downlink channel state information (CSI) used for precoding is conveyed back in massive multi-input multi-output (MIMO) systems with frequency division duplex (FDD) mode, which costs too much bandwidth. However, most of CSI compression algorithms including deep learning based schemes show poor performance with high compression ratio (CR) in ourdoor scenery. In this letter, we propose a new neural network that utilizes the deep neural network (DNN) to improve the learning ability of residual network and a novel activation function to adapt the neural network. Simulation results show that the proposed CsiNet+DNN method exhibits better performance than exsiting methods in recovery quality and accuracy, which also achieves remarkable robustness at noise situations.

Randomized Low-Rank Approximation Based Massive MIMO CSI Compression

Massive multiple-input multiple-output (MIMO) is regarded as one enabling technique to improve channel capacity and energy/spectrum efficiency of 5G communications. To attain such potential benefits, accurate channel information is critical to the transmitter, which yet remains a challenging task for frequency division duplexing (FDD) systems, i.e., the channel state information (CSI) feedback tends to be resource-demanding especially for massive MIMO communications. In this work, we propose a novel CSI feedback method with low complexity and high precision, which is inspired by randomized matrix approximation. Our approach exploits the inherent low-rank characteristic of a large channel matrix, and approximates it by small sub-matrices which are then reported to transmitter to recover a CSI matrix. Theoretical bounds of the recovered CSI in both error-free and error cases are derived. Simulation results demonstrate our method could recover CSI accurately via an extremely low complexity and yet achieve a largely reduced compression ratio (or feedback overhead), compared to other schemes. It thus has the great potential in the emerging massive MIMO FDD communications.

A Lightweight Deep Network for Efficient CSI Feedback in Massive MIMO Systems

To fully exploit the advantages of massive multiple-input multiple-output (m-MIMO), accurate channel state information (CSI) is required at the transmitter. However, excessive CSI feedback for large antenna arrays is inefficient and thus undesirable in practical applications. By exploiting the inherent correlation characteristics of complex-valued channel responses in the angular-delay domain, we propose a novel neural network (NN) architecture, namely ENet, for CSI compression and feedback in m-MIMO. Even if the ENet processes the real and imaginary parts of the CSI values separately, its special structure enables the network trained for the real part only to be reused for the imaginary part. The proposed ENet shows enhanced performance with the network size reduced by nearly an order of magnitude compared to the existing NN-based solutions. Experimental results verify the effectiveness of the proposed ENet.

Binary Neural Network Aided CSI Feedback in Massive MIMO System

In massive multiple-input multiple-output (MIMO) system, channel state information (CSI) is essential for the base station (BS) to achieve high performance gain. The CSI matrix needs to be estimated and fed back from user equipment (UE) in frequency division duplexing (FDD) mode. Recently, deep learning is widely used in CSI compression to reduce the feedback overhead. However, applying neural network brings extra memory and computation cost, which is non-negligible especially for the resource limited UE. In this letter, a novel binarization aided feedback network named BCsiNet is introduced to lighten the encoder at UE. The proposed BCsiNet offers over 30× memory saving and around 2× inference acceleration for the encoder compared with CsiNet. Moreover, experiments show that the feedback performance of BCsiNet can be comparable with original CsiNet even with the encoder binarized.