Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

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

Jiajia Guo, Chao-Kai Wen, Shi Jin, Geoffrey Ye Li

TL;DR

A comprehensive overview of state-of-the-art research on deep learning-based CSI feedback is provided, beginning with basic DL concepts widely used in CSI feedback and then categorizing and describing some existing DL-based feedback works.

Abstract

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 terminal) and feeding back to the transmitter. The overhead of CSI feedback occupies substantial uplink bandwidth resources, especially when the number of the 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 bit-level CSI feedback and joint design of CSI feedback with other communication modules are also introduced, and some practical issues, including training dataset collection, online training, complexity, generalization, 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.

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

TL;DR

A comprehensive overview of state-of-the-art research on deep learning-based CSI feedback is provided, beginning with basic DL concepts widely used in CSI feedback and then categorizing and describing some existing DL-based feedback works.

Abstract

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 terminal) and feeding back to the transmitter. The overhead of CSI feedback occupies substantial uplink bandwidth resources, especially when the number of the 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 bit-level CSI feedback and joint design of CSI feedback with other communication modules are also introduced, and some practical issues, including training dataset collection, online training, complexity, generalization, 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.
Paper Structure (57 sections, 26 equations, 33 figures, 6 tables)

This paper contains 57 sections, 26 equations, 33 figures, 6 tables.

Table of Contents

  1. Introduction
  2. Conventional feedback schemes
  3. Massive MIMO System
  4. Codebook-based CSI Feedback
  5. CS-based CSI Feedback
  6. Common Deep Network Architectures
  7. FC Layer
  8. Convolutional Layer
  9. LSTMs
  10. VAE
  11. GAN
  12. Attention Mechanism
  13. DL-based CSI Feedback
  14. Novel NN Architecture Design
  15. Increasing Receptive Field
  16. ...and 42 more sections

Figures (33)

  • Figure 1: Illustration of autoencoder architectures. In image compression, the NN-based encoder compresses the original image into a low-dimensional representation and then the NN-based decoder reconstructs the image from the latent representation. The encoder and decoder are jointly trained. In the right sub-figure, the downlink CSI is regarded as a special type of "image".
  • Figure 2: Outline of article
  • Figure 3: Illustration of codebook-based CSI feedback. The codebook is known to the user and the BS. The user searches the codeword, which is the closest to the downlink CSI, and feeds back the corresponding index to the BS. Upon receiving the index, the BS can obtain the channel by looking up the shared codebook.
  • Figure 4: Receptive field illustration of two stacked $3\times 3$ convolutional layers. Stride $s$ is set as 1. The upper "pixel" is determined by the $3\times 3$ square area in the middle. Each intermediate "pixel" is determined by the input $3\times 3$ square area, which is overlapped with one another. Therefore, the upper "pixel" is determined by the $5\times 5$ square area of the input.
  • Figure 5: Illustration of an LSTM cell Understanding that has feedback connections and allows information to persist
  • ...and 28 more figures