Multi-resolution CSI Feedback with Deep Learning in Massive MIMO System

Lu, Zhilin; Wang, Jintao; Song, Jian

doi:10.1109/icc40277.2020.9149229

Cited by 173 publications

(139 citation statements)

References 16 publications

Supporting

Mentioning

139

Contrasting

Order By: Relevance

“…A “deep complex-valued convolutional network” that does not rely explicitly on the Fourier transform is designed in [ 85 ] to recover bits from time-domain orthogonal frequency-division multiplexing signals. A novel feedback network CRNet is presented in [ 86 ] using advanced training techniques to get superior performance by extracting CSI features on multiple resolutions.…”

Section: Rl and DL Application In Mimomentioning

confidence: 99%

Application of Reinforcement Learning and Deep Learning in Multiple-Input and Multiple-Output (MIMO) Systems

Naeem

Pietro

Coronato

2021

Sensors

View full text Add to dashboard Cite

The current wireless communication infrastructure has to face exponential development in mobile traffic size, which demands high data rate, reliability, and low latency. MIMO systems and their variants (i.e., Multi-User MIMO and Massive MIMO) are the most promising 5G wireless communication systems technology due to their high system throughput and data rate. However, the most significant challenges in MIMO communication are substantial problems in exploiting the multiple-antenna and computational complexity. The recent success of RL and DL introduces novel and powerful tools that mitigate issues in MIMO communication systems. This article focuses on RL and DL techniques for MIMO systems by presenting a comprehensive review on the integration between the two areas. We first briefly provide the necessary background to RL, DL, and MIMO. Second, potential RL and DL applications for different MIMO issues, such as detection, classification, and compression; channel estimation; positioning, sensing, and localization; CSI acquisition and feedback, security, and robustness; mmWave communication and resource allocation, are presented.

show abstract

Section: Rl and DL Application In Mimomentioning

confidence: 99%

Application of Reinforcement Learning and Deep Learning in Multiple-Input and Multiple-Output (MIMO) Systems

Naeem

Pietro

Coronato

2021

Sensors

View full text Add to dashboard Cite

show abstract

“…We compare the performance of PRVNet with three stateof-the-art CS-based methods, namely, Lasso L 1 -solver [16], TVAL-3 [17], and BM3D-AMP [18]. In addition, we compare the performance of our proposed model to two recent deep learning-based methods, namely, CsiNet [5] and CRNet [7]. In order to evaluate the performance of different methods, we measure the distance between the original CSI matrix, H a , and the reconstruction image,Ĥ a , by means of normalized mean square error: Table I shows the performance of the proposed PRVNet against the state-of-the-art methods.…”

Section: B Performance Of Prvnetmentioning

confidence: 99%

“…In [7], the authors proposed a neural network architecture, called CRNet, for multi-resolution CSI feedback in massive MIMO. The model is shown to have an improved performance against classic CS-based techniques as well as CsiNet.…”

Section: Introductionmentioning

confidence: 99%

PRVNet: Variational Autoencoders for Massive MIMO CSI Feedback

Hussien¹

2020

Preprint

View full text Add to dashboard Cite

In a frequency division duplexing multiple-input multiple-output (FDD-MIMO) system, the user equipment (UE) send the downlink channel state information (CSI) to the base station for performance improvement. However, with the growing complexity of MIMO systems, this feedback becomes expensive and has a negative impact on the bandwidth. Although this problem has been largely studied in the literature, the noisy nature of the feedback channel is less considered. In this paper, we introduce PRVNet, a neural architecture based on variational autoencoders (VAE). VAE gained large attention in many fields (e.g., image processing, language models, or recommendation system). However, it received less attention in the communication domain generally and in CSI feedback problem specifically. We also introduce a different regularization parameter for the learning objective, which proved to be crucial for achieving competitive performance. In addition, we provide an efficient way to tune this parameter using KL-annealing. Empirically, we show that the proposed model significantly outperforms state-of-the-art, including two neural network approaches. The proposed model is also proved to be more robust against different levels of noise.

show abstract

“…CsiNet [3] provides a fundamental structure for the deep learning-based CSI feedback research. Apart from the optimization for feedback accuracy [4][5][6][7], there are some researches that focus on the feasibility of the feedback network. [8] and [9] have achieved the optimization of the model complexity at the UE from the perspectives of compression rate and data type of the parameter.…”

mentioning

confidence: 99%

“…The proposed architecture can be applied to the existing (or future) CSI feedback networks with convolutional layer and fully connected (FC) layer at the encoder. Note that some previous works [10,11] omit the FC layers from the network with a cost of network complexity, most exiting researches apply the FC layer in the encoder, for example, CsiNet [3], CRNet [7] and CsiNet+ [8]. Specific training procedure of the FCS-network is provided and comparable performance can be obtained with great reduction of the parameters at the UE compared with the original network which uses different models in different scenarios.…”

mentioning

confidence: 99%

Fully connected layer‐shared network architecture for massive MIMO CSI feedback

Zhang

Liang

et al. 2022

Electronics Letters

View full text Add to dashboard Cite

In frequency division duplex (FDD) massive multiple-input multipleoutput (MIMO) system, deep learning-based channel state information (CSI) feedback approach improves CSI reconstruction accuracy. However, different models need to be trained and stored for different scenarios, which causes huge memory cost, especially at the user equipment (UE) side. In this letter, a fully connected layer-shared (FCS) feedback architecture is proposed to simplify the feedback network at the UE. As a case study, the FCS architecture is applied on CsiNet and CRNet in the indoor and outdoor scenarios. Experimental results indicate that the network using the FCS architecture is able to achieve comprehensive performance with impressive storage saving.

show abstract

Multi-resolution CSI Feedback with Deep Learning in Massive MIMO System

Cited by 173 publications

References 16 publications

Application of Reinforcement Learning and Deep Learning in Multiple-Input and Multiple-Output (MIMO) Systems

Application of Reinforcement Learning and Deep Learning in Multiple-Input and Multiple-Output (MIMO) Systems

PRVNet: Variational Autoencoders for Massive MIMO CSI Feedback

Fully connected layer‐shared network architecture for massive MIMO CSI feedback

Contact Info

Product

Resources

About