Zirui Chen
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Zirui Chen has a total of 27 co-authors. Here are the top co-authors by number of shared publications:
- Zhaoyang Zhang
Zhejiang University
11 shared publications - Zhaohui Yang
Zhejiang University, Hangzhou, China
9 shared publications - Chongwen Huang
Zhejiang University
3 shared publications - Richeng Jin
Department of Information and Communication Engineering, Zhejiang University, Hangzhou, China
3 shared publications - Zhuoran Xiao
Nokia Bell Labs, Shanghai, China
3 shared publications - Kai-Kit Wong
University College London
2 shared publications
Latest Publications
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25 (11) , 21312-21329
AC$^{2}$P$^{2}$SL: Adaptive Communication-Computation Pipeline Parallel Split Learning Over Edge Networks
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25 (11) , 21312-21329
AC$^{2}$P$^{2}$SL: Adaptive Communication-Computation Pipeline Parallel Split Learning Over Edge Networks
- 25 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 25 (15361276) , 12691-12708
Multi-View Wireless Sensing via Conditional Generative Learning: Framework and Model Design
In this paper, we incorporate physical knowledge into learning-based high-precision target sensing using the multi-view channel state information (CSI) between multiple base stations (BSs) and user equipment (UEs). Such kind of multi-view sensing problem can be naturally cast into a conditional generation framework. To this end, we design a bipartite neural network architecture, the first part of which uses an elaborately designed encoder to fuse the latent target features embedded in the multi-view CSI, and then the second uses them as conditioning inputs of a powerful generative model to guide the target’s reconstruction. Specifically, the encoder is designed to capture the physical correlation between the CSI and the target, and also be adaptive to the numbers and positions of BS-UE pairs. Therein the view-specific nature of CSI is assimilated by introducing a spatial positional embedding scheme, which exploits the structure of electromagnetic(EM)-wave propagation channels. Finally, a conditional diffusion model with a weighted loss is employed to generate the target’s point cloud from the fused features. Extensive numerical results demonstrate that the proposed generative multi-view (Gen-MV) sensing framework exhibits excellent flexibility and significant performance improvement on the reconstruction quality of target’s shape and EM properties.
- 25 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 25 (15361276) , 12691-12708
Multi-View Wireless Sensing via Conditional Generative Learning: Framework and Model Design
In this paper, we incorporate physical knowledge into learning-based high-precision target sensing using the multi-view channel state information (CSI) between multiple base stations (BSs) and user equipment (UEs). Such kind of multi-view sensing problem can be naturally cast into a conditional generation framework. To this end, we design a bipartite neural network architecture, the first part of which uses an elaborately designed encoder to fuse the latent target features embedded in the multi-view CSI, and then the second uses them as conditioning inputs of a powerful generative model to guide the target’s reconstruction. Specifically, the encoder is designed to capture the physical correlation between the CSI and the target, and also be adaptive to the numbers and positions of BS-UE pairs. Therein the view-specific nature of CSI is assimilated by introducing a spatial positional embedding scheme, which exploits the structure of electromagnetic(EM)-wave propagation channels. Finally, a conditional diffusion model with a weighted loss is employed to generate the target’s point cloud from the fused features. Extensive numerical results demonstrate that the proposed generative multi-view (Gen-MV) sensing framework exhibits excellent flexibility and significant performance improvement on the reconstruction quality of target’s shape and EM properties.
- 26 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 24 (8) , 6365-6381
A Hybrid Inference Architecture Incorporating Neural Network With Belief Propagation for AI Receivers
Conventional wireless communication receivers guided by Bayesian inference methods need to know the exact statistical relationship among variables, which is hard to obtain accurately in wireless contexts, thus limiting the system performance. The recently emerging artificial intelligence (AI)-empowered algorithms have shown striking performances in exploring the implicit relationship among variables with specially designed Neural Networks (NNs). Therefore, it is preferable to integrate NNs with BPs in receiver design. Such approaches also leverage NNs’ lack of reasoning ability in large state spaces and traditional BPs’ lack of reasoning depth. However, conventional receiver modules are usually designed based on explicit mathematical derivations, which cannot be easily substituted with datadriven NNs as they may break the overall inner relationship of the algorithm. In this paper, we investigate how to beneficially incorporate NNs into the existing Belief Propagation (BP)-based framework, taking the traditional semi-blind estimation problem in an Orthogonal Frequency-Division Multiplexing (OFDM) receiver as an example. Unlike existing deep-unfolding approaches, we simply utilize NNs as embedded functional units rather than duplicate denoising modules. Through qualitative discussions and numerical results, we illustrate the characteristics, principles, and differences of our proposed architecture compared to the traditional BP framework and show the dramatic performance improvements brought by incorporating NNs with BP in this well-investigated problem. Recalling that the state evolution of NNs is different from that of traditional BP methods, we give some new insights and design principles which are somehow counterfactual. We also raise some open issues on the incorporated framework.
- 26 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 24 (8) , 6365-6381
A Hybrid Inference Architecture Incorporating Neural Network With Belief Propagation for AI Receivers
Conventional wireless communication receivers guided by Bayesian inference methods need to know the exact statistical relationship among variables, which is hard to obtain accurately in wireless contexts, thus limiting the system performance. The recently emerging artificial intelligence (AI)-empowered algorithms have shown striking performances in exploring the implicit relationship among variables with specially designed Neural Networks (NNs). Therefore, it is preferable to integrate NNs with BPs in receiver design. Such approaches also leverage NNs’ lack of reasoning ability in large state spaces and traditional BPs’ lack of reasoning depth. However, conventional receiver modules are usually designed based on explicit mathematical derivations, which cannot be easily substituted with datadriven NNs as they may break the overall inner relationship of the algorithm. In this paper, we investigate how to beneficially incorporate NNs into the existing Belief Propagation (BP)-based framework, taking the traditional semi-blind estimation problem in an Orthogonal Frequency-Division Multiplexing (OFDM) receiver as an example. Unlike existing deep-unfolding approaches, we simply utilize NNs as embedded functional units rather than duplicate denoising modules. Through qualitative discussions and numerical results, we illustrate the characteristics, principles, and differences of our proposed architecture compared to the traditional BP framework and show the dramatic performance improvements brought by incorporating NNs with BP in this well-investigated problem. Recalling that the state evolution of NNs is different from that of traditional BP methods, we give some new insights and design principles which are somehow counterfactual. We also raise some open issues on the incorporated framework.
- 20 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-16
VBIM-Net: Variational Born Iterative Network for Inverse Scattering Problems
Recently, studies have shown the potential of integrating field-type iterative methods with deep learning (DL) techniques in solving inverse scattering problems (ISPs). In this article, we propose a novel Variational Born Iterative Network, namely, VBIM-Net, to solve the full-wave ISPs with significantly improved structural rationality and inversion quality. The proposed VBIM-Net emulates the alternating updates of the total electric field and the contrast in the variational Born iterative method (VBIM) by multiple layers of subnetworks. We embed the analytical calculation of the contrast variation into each subnetwork, converting the scattered field residual into an approximate contrast variation and then enhancing it by a U-Net, thus avoiding the requirement of matched measurement dimension and grid resolution as in existing approaches. The total field and contrast of each layer’s output is supervised in the loss function of VBIM-Net, imposing soft physical constraints on the variables in the subnetworks, which benefits the model’s performance. In addition, we design a training scheme with extra noise to enhance the model’s stability. Extensive numerical results on synthetic and experimental data both verify the inversion quality, generalization ability, and robustness of the proposed VBIM-Net. This work may provide some new inspiration for the design of efficient field-type DL schemes.
- 20 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-16
VBIM-Net: Variational Born Iterative Network for Inverse Scattering Problems
Recently, studies have shown the potential of integrating field-type iterative methods with deep learning (DL) techniques in solving inverse scattering problems (ISPs). In this article, we propose a novel Variational Born Iterative Network, namely, VBIM-Net, to solve the full-wave ISPs with significantly improved structural rationality and inversion quality. The proposed VBIM-Net emulates the alternating updates of the total electric field and the contrast in the variational Born iterative method (VBIM) by multiple layers of subnetworks. We embed the analytical calculation of the contrast variation into each subnetwork, converting the scattered field residual into an approximate contrast variation and then enhancing it by a U-Net, thus avoiding the requirement of matched measurement dimension and grid resolution as in existing approaches. The total field and contrast of each layer’s output is supervised in the loss function of VBIM-Net, imposing soft physical constraints on the variables in the subnetworks, which benefits the model’s performance. In addition, we design a training scheme with extra noise to enhance the model’s stability. Extensive numerical results on synthetic and experimental data both verify the inversion quality, generalization ability, and robustness of the proposed VBIM-Net. This work may provide some new inspiration for the design of efficient field-type DL schemes.
- 20 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Selected Areas in Communications
- Vol. 43 (3) , 944-958
Channel Deduction: A New Learning Framework to Acquire Channel From Outdated Samples and Coarse Estimate
How to reduce the pilot overhead required for channel estimation? How to deal with the channel dynamic changes and error propagation in channel prediction? To jointly address these two critical issues in next-generation transceiver design, in this paper, we propose a novel framework named channel deduction for high-dimensional channel acquisition in multiple-input multiple-output (MIMO)-orthogonal frequency division multiplexing (OFDM) systems. Specifically, it makes use of the outdated channel information of past time slots, performs coarse estimation for the current channel with a relatively small number of pilots, and then fuses these two information to obtain a complete representation of the present channel. The rationale is to align the current channel representation to both the latent channel features within the past samples and the coarse estimate of current channel at the pilots, which, in a sense, behaves as a complementary combination of estimation and prediction and thus reduces the overall overhead. To fully exploit the highly nonlinear correlations in time, space, and frequency domains, we resort to learning-based implementation approaches. By using the highly efficient complex-domain multilayer perceptron (MLP)-mixer for across-space-frequency-domain representation and the recurrence-based or attention-based mechanisms for the past-present interaction, we respectively design two different channel deduction neural networks (CDNets). We provide a general procedure of data collection, training, and deployment to standardize the application of CDNets. Comprehensive experimental evaluations in accuracy, robustness, and efficiency demonstrate the superiority of the proposed approach, which reduces the pilot overhead by up to 88.9% compared to state-of-the-art estimation approaches and enables continuous operating even under unknown user movement and error propagation.
- 20 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Selected Areas in Communications
- Vol. 43 (3) , 944-958
Channel Deduction: A New Learning Framework to Acquire Channel From Outdated Samples and Coarse Estimate
How to reduce the pilot overhead required for channel estimation? How to deal with the channel dynamic changes and error propagation in channel prediction? To jointly address these two critical issues in next-generation transceiver design, in this paper, we propose a novel framework named channel deduction for high-dimensional channel acquisition in multiple-input multiple-output (MIMO)-orthogonal frequency division multiplexing (OFDM) systems. Specifically, it makes use of the outdated channel information of past time slots, performs coarse estimation for the current channel with a relatively small number of pilots, and then fuses these two information to obtain a complete representation of the present channel. The rationale is to align the current channel representation to both the latent channel features within the past samples and the coarse estimate of current channel at the pilots, which, in a sense, behaves as a complementary combination of estimation and prediction and thus reduces the overall overhead. To fully exploit the highly nonlinear correlations in time, space, and frequency domains, we resort to learning-based implementation approaches. By using the highly efficient complex-domain multilayer perceptron (MLP)-mixer for across-space-frequency-domain representation and the recurrence-based or attention-based mechanisms for the past-present interaction, we respectively design two different channel deduction neural networks (CDNets). We provide a general procedure of data collection, training, and deployment to standardize the application of CDNets. Comprehensive experimental evaluations in accuracy, robustness, and efficiency demonstrate the superiority of the proposed approach, which reduces the pilot overhead by up to 88.9% compared to state-of-the-art estimation approaches and enables continuous operating even under unknown user movement and error propagation.
- 28 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Network
- Vol. 38 (5) , 37-46
An Edge-Cloud Collaboration Framework for Generative AI Service Provision With Synergetic Big Cloud Model and Small Edge Models
Generative artificial intelligence (GenAI) offers various services to users through content creation, which is believed to be one of the most important components in future networks. However, training and deploying big artificial intelligence models (BAIMs) introduces substantial computational and communication overhead. This poses a critical challenge to centralized approaches, due to the need of high-performance computing infrastructure and the reliability, secrecy and timeliness issues in long-distance access of cloud services. Therefore, there is an urging need to decentralize the services, partly moving them from the cloud to the edge and establishing native GenAI services to enable private, timely, and personalized experiences. In this paper, we propose a brand-new bottom-up BAIM architecture with synergetic big cloud model and small edge models, and design a distributed training framework and a task-oriented deployment scheme for efficient provision of native GenAI services. The proposed framework can facilitate collaborative intelligence, enhance adaptability, gather edge knowledge and alleviate edge-cloud burden. The effectiveness of the proposed framework is demonstrated through an image generation use case. Finally, we outline fundamental research directions to fully exploit the collaborative potential of edge and cloud for native GenAI and BAIM applications.
- 28 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Network
- Vol. 38 (5) , 37-46
An Edge-Cloud Collaboration Framework for Generative AI Service Provision With Synergetic Big Cloud Model and Small Edge Models
Generative artificial intelligence (GenAI) offers various services to users through content creation, which is believed to be one of the most important components in future networks. However, training and deploying big artificial intelligence models (BAIMs) introduces substantial computational and communication overhead. This poses a critical challenge to centralized approaches, due to the need of high-performance computing infrastructure and the reliability, secrecy and timeliness issues in long-distance access of cloud services. Therefore, there is an urging need to decentralize the services, partly moving them from the cloud to the edge and establishing native GenAI services to enable private, timely, and personalized experiences. In this paper, we propose a brand-new bottom-up BAIM architecture with synergetic big cloud model and small edge models, and design a distributed training framework and a task-oriented deployment scheme for efficient provision of native GenAI services. The proposed framework can facilitate collaborative intelligence, enhance adaptability, gather edge knowledge and alleviate edge-cloud burden. The effectiveness of the proposed framework is demonstrated through an image generation use case. Finally, we outline fundamental research directions to fully exploit the collaborative potential of edge and cloud for native GenAI and BAIM applications.
- 26 February 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Wireless Communications Letters
- Vol. 13 (5) , 1369-1373
Channel Mapping Based on Interleaved Learning With Complex-Domain MLP-Mixer
In multiple-input multiple-output (MIMO) orthogonal frequency division multiplexing (OFDM) systems, representing the whole channel only based on partial subchannels will significantly reduce the channel acquisition overhead. For such a channel mapping task, inspired by the intrinsic coupling across the space and frequency domains, this letter proposes to use interleaved learning with partial antenna and subcarrier characteristics to represent the whole MIMO-OFDM channel. Specifically, we design a complex-domain multilayer perceptron (MLP)-Mixer (CMixer), which utilizes two kinds of complex-domain MLP modules to learn the space and frequency channel characteristics respectively and then couple the learned properties by interleaving. The complex-domain computation well preserves the specific intra-domain channel structures, while the interleaving facilitates the inter-domain knowledge exchanges. The resultant physics-inspired CMixer greatly reduces the learning burden, exhibiting extremely high efficiency on channel mapping than existing data-driven approaches. Simulation shows that under typical scenarios and with only light-weighted networks, it brings 4.610dB gains in normalized channel error and up to 67.3% channel data reduction compared to existing schemes while keeping better than 95% cosine correlations w.r.t. the original channels.
- 26 February 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Wireless Communications Letters
- Vol. 13 (5) , 1369-1373
Channel Mapping Based on Interleaved Learning With Complex-Domain MLP-Mixer
In multiple-input multiple-output (MIMO) orthogonal frequency division multiplexing (OFDM) systems, representing the whole channel only based on partial subchannels will significantly reduce the channel acquisition overhead. For such a channel mapping task, inspired by the intrinsic coupling across the space and frequency domains, this letter proposes to use interleaved learning with partial antenna and subcarrier characteristics to represent the whole MIMO-OFDM channel. Specifically, we design a complex-domain multilayer perceptron (MLP)-Mixer (CMixer), which utilizes two kinds of complex-domain MLP modules to learn the space and frequency channel characteristics respectively and then couple the learned properties by interleaving. The complex-domain computation well preserves the specific intra-domain channel structures, while the interleaving facilitates the inter-domain knowledge exchanges. The resultant physics-inspired CMixer greatly reduces the learning burden, exhibiting extremely high efficiency on channel mapping than existing data-driven approaches. Simulation shows that under typical scenarios and with only light-weighted networks, it brings 4.610dB gains in normalized channel error and up to 67.3% channel data reduction compared to existing schemes while keeping better than 95% cosine correlations w.r.t. the original channels.
- 8 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 23 (7) , 7186-7199
From Data-Driven Learning to Physics-Inspired Inferring: A Novel Mobile MIMO Channel Prediction Scheme Based on Neural ODE
In this paper, we propose an innovative learning-based channel prediction scheme so as to achieve higher prediction accuracy and reduce the requirements of huge amounts and strict sequential format of channel data. Inspired by the idea of the neural ordinary differential equation (Neural ODE), we first prove that the channel prediction problem can be modeled as an ODE problem with a known initial value by analyzing the physical process of electromagnetic wave propagation within a mobile environment. Then, we design a novel physics-inspired spatial channel gradient network (SCGnet), which represents the derivative process of channel varying as a special neural network and can obtain the gradients at any relative displacement needed for the ODE solving. With the SCGnet, the static channel at any location served by the base station is accurately inferred through consecutive propagation and integration. Finally, we design an efficient recurrent positioning algorithm based on some prior knowledge of user mobility to obtain the velocity vector and propose an approximate Doppler compensation method to make up the instantaneous angular-delay domain channel. Only discrete historical channel data is needed for the training, whereas only a few fresh channel measurements are needed for the prediction, which ensures the scheme’s practicability. Comprehensive evaluations show that the proposed scheme is most efficient in representing, learning, and predicting mobile wireless channels.
- 8 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 23 (7) , 7186-7199
From Data-Driven Learning to Physics-Inspired Inferring: A Novel Mobile MIMO Channel Prediction Scheme Based on Neural ODE
In this paper, we propose an innovative learning-based channel prediction scheme so as to achieve higher prediction accuracy and reduce the requirements of huge amounts and strict sequential format of channel data. Inspired by the idea of the neural ordinary differential equation (Neural ODE), we first prove that the channel prediction problem can be modeled as an ODE problem with a known initial value by analyzing the physical process of electromagnetic wave propagation within a mobile environment. Then, we design a novel physics-inspired spatial channel gradient network (SCGnet), which represents the derivative process of channel varying as a special neural network and can obtain the gradients at any relative displacement needed for the ODE solving. With the SCGnet, the static channel at any location served by the base station is accurately inferred through consecutive propagation and integration. Finally, we design an efficient recurrent positioning algorithm based on some prior knowledge of user mobility to obtain the velocity vector and propose an approximate Doppler compensation method to make up the instantaneous angular-delay domain channel. Only discrete historical channel data is needed for the training, whereas only a few fresh channel measurements are needed for the prediction, which ensures the scheme’s practicability. Comprehensive evaluations show that the proposed scheme is most efficient in representing, learning, and predicting mobile wireless channels.
- 9 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Selected Areas in Communications
- Vol. 41 (12) , 3848-3862
Deep Learning-Based Multi-User Positioning in Wireless FDMA Cellular Networks
In Cooperative Intelligent Transportation Systems (C-ITS) and Connected Automated Vehicles (CAV), accessing multiple users and providing high-precision positioning are both vital. This paper aims to design an efficient deep learning approach to extend current Channel State Information (CSI)-based positioning to Frequency Division Multiple Access (FDMA) mode. In FDMA mode, different users are allocated with different subcarriers, making the user CSI have diverse frequency domain characteristics. The diverse frequency domain characteristics bring huge interference to the neural network for stable position inference, and efficient designs are required to handle this challenge. This paper proposes a novel approach named multi-frequency fusion learning for CSI-based positioning. By first using a shareable method to extract position-related features from CSI on each subcarrier independently and then fusing the obtained features, the designed neural network obtains excellent frequency domain flexibility to cope with the diverse frequency address challenge in FDMA mode. Meanwhile, we provide the feasibility analysis of this learning approach in massive Multiple-Input Multiple-Output (MIMO) systems to ensure its stable application. Based on the architecture of multi-frequency fusion learning, we propose two specific positioning schemes with differentiated designs. One is a Multi-Frequency Ensemble Network (MFENet), which extracts and fuses frequency-independent features to ensure the network is utterly unharmed by the complicated frequency domain characteristics. The other is a Multi-Frequency Cumulative Network (MFCNet), which uses sufficient feature accumulation to achieve high precision positioning. The key performance indices and applications on vehicles are comprehensively compared with popular deep-learning methods. Experiment results show the effectiveness and superiority of the proposed schemes.
- 9 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Selected Areas in Communications
- Vol. 41 (12) , 3848-3862
Deep Learning-Based Multi-User Positioning in Wireless FDMA Cellular Networks
In Cooperative Intelligent Transportation Systems (C-ITS) and Connected Automated Vehicles (CAV), accessing multiple users and providing high-precision positioning are both vital. This paper aims to design an efficient deep learning approach to extend current Channel State Information (CSI)-based positioning to Frequency Division Multiple Access (FDMA) mode. In FDMA mode, different users are allocated with different subcarriers, making the user CSI have diverse frequency domain characteristics. The diverse frequency domain characteristics bring huge interference to the neural network for stable position inference, and efficient designs are required to handle this challenge. This paper proposes a novel approach named multi-frequency fusion learning for CSI-based positioning. By first using a shareable method to extract position-related features from CSI on each subcarrier independently and then fusing the obtained features, the designed neural network obtains excellent frequency domain flexibility to cope with the diverse frequency address challenge in FDMA mode. Meanwhile, we provide the feasibility analysis of this learning approach in massive Multiple-Input Multiple-Output (MIMO) systems to ensure its stable application. Based on the architecture of multi-frequency fusion learning, we propose two specific positioning schemes with differentiated designs. One is a Multi-Frequency Ensemble Network (MFENet), which extracts and fuses frequency-independent features to ensure the network is utterly unharmed by the complicated frequency domain characteristics. The other is a Multi-Frequency Cumulative Network (MFCNet), which uses sufficient feature accumulation to achieve high precision positioning. The key performance indices and applications on vehicles are comprehensively compared with popular deep-learning methods. Experiment results show the effectiveness and superiority of the proposed schemes.
- 6 March 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 22 (11) , 7393-7407
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.
- 6 March 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 22 (11) , 7393-7407
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.