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

Sun, Yan; Xu, Wei; Liang, Le; Wang, Ning; Li, Geoffery Ye; You, Xiaohu

doi:10.1109/lwc.2021.3083331

Cited by 38 publications

(20 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We stress that each of the encoder-decoder network pair shares the same network architecture and parameters. Unlike existing neural network based techniques for CSI compression, e.g., in [15], [16], the proposed DisNet is based on a distributed neural network architecture which is therefore suitable for deployment in multiuser mmWave MIMO. In the DisNet, each encoder-decoder network pair works for one UE.…”

Section: A Disnet For Limited Feedbackmentioning

confidence: 99%

Distributed Neural Precoding for Hybrid mmWave MIMO Communications With Limited Feedback

Wei

et al. 2022

IEEE Commun. Lett.

Self Cite

View full text Add to dashboard Cite

Hybrid precoding is a cost-efficient technique for millimeter wave (mmWave) massive multiple-input multipleoutput (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 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.

show abstract

Section: A Disnet For Limited Feedbackmentioning

confidence: 99%

Distributed Neural Precoding for Hybrid mmWave MIMO Communications With Limited Feedback

Wei

et al. 2022

IEEE Commun. Lett.

Self Cite

View full text Add to dashboard Cite

show abstract

“…To improve the accuracy of CSI feedback, a DL-based network named CRNet [27] was proposed to achieve better performance via extracting CSI features on multiple resolutions. Besides, in [28], a neural network named ENet was trained for only the real part of CSI by exploiting the inherent correlation characteristics between the real and imaginary parts of complex-valued channel responses. Also in [29], a neural network named CLNet was proposed to utilize a forged complex-valued input layer to process signals and the spatial-attention to enhance the performance.…”

Section: A Related Workmentioning

confidence: 99%

Deep CSI Compression for Massive MIMO: A Self-information Model-driven Neural Network

Yin¹,

Wang²,

Xie³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

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 modeland-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.

show abstract

Section: Multi-rate Csi Feedback Framework With Flexible Number Of An...mentioning

confidence: 99%

“…There have been notable progresses in terms of recovery performance among the recent autoencoder-based CSI feedback frameworks [8], [11], [17], [18]. Since UEs have limited resources [18], an important consideration is the reduction of complexity and required storage at UE. Unfortunately, naïvely following the example of autoencoders in image compression leads to the directly input of full DL CSI matrix H as an image to deep learning architecture to extract pixel-wise features.…”

Section: Multi-rate Csi Feedback Framework With Flexible Number Of An...mentioning

confidence: 99%

“…We first define a new quantity, base number, to be the dimension of the spatial domain of the new framework input. Assuming base number is K, as compared to the full DL CSI with size of N a × N f , we first divide the full DL CSI matrix into N a /K secondary DL CSI matrices with the same size of K × N f given as follows: With the knowledge of similar statistics of real and imaginary parts of CSIs [18], we can further decompose the DL CSI into real and imaginary parts and feed them into an autoencoder network individually for DL pilot CSI compression and recovery which can be expressed as…”

Section: A Divide-and-conquer (Dc) Frameworkmentioning

confidence: 99%

See 1 more Smart Citation

A Scalable Deep Learning Framework for Multi-rate CSI Feedback under Variable Antenna Ports

Lin¹,

Lee²,

Ding³

2022

Preprint

View full text Add to dashboard Cite

Channel state information (CSI) at transmitter is crucial for massive MIMO downlink systems to achieve high spectrum and energy efficiency. Existing works have provided deep learning architectures for CSI feedback and recovery at the eNB/gNB by reducing user feedback overhead and improving recovery accuracy. However, existing DL architectures tend to be inflexible and non-scalable as models are often trained according to a preset number of antennas for a given compression ratio. In this work, we develop a flexible and scalable learning frame work based on a divide-and-conquer approach (DCA). This new DCA architecture can flexibly accommodate different numbers of 3GPP antenna ports and dynamic levels of feedback compression. Importantly, it also significantly reduces computational complexity and memory size by allowing UEs to feedback segmented downlink CSI. We further propose a multi-rate successive convolution encoder with fewer than 1000 parameters. Test results demonstrate superior performance, good scalability, and low complexity for both indoor and outdoor channels.

show abstract

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

Cited by 38 publications

References 26 publications

Distributed Neural Precoding for Hybrid mmWave MIMO Communications With Limited Feedback

Distributed Neural Precoding for Hybrid mmWave MIMO Communications With Limited Feedback

Deep CSI Compression for Massive MIMO: A Self-information Model-driven Neural Network

A Scalable Deep Learning Framework for Multi-rate CSI Feedback under Variable Antenna Ports

Contact Info

Product

Resources

About