Deep Learning for Massive MIMO CSI Feedback

By Chao-Kai Wen, Wan-Ting Shih, Shi Jin. IEEE Wireless Communications Letters, 2017

1,134 citations recorded by the source corpus.

In frequency division duplex mode, the downlink channel state information (CSI) should be sent to the base station through feedback links so that the potential gains of a massive multiple-input multiple-output can be exhibited. However, such a transmission is hindered by excessive feedback overhead. In this letter, we use deep learning technology to develop CsiNet, a novel CSI sensing and recovery mechanism that learns to effectively use channel structure from training samples. CsiNet learns a transformation from CSI to a near-optimal number of representations (or codewords) and an inverse transformation from codewords to CSI. We perform experiments to demonstrate that CsiNet can recover CSI with significantly improved reconstruction quality compared with existing compressive sensing (CS)-based methods. Even at excessively low compression regions where CS-based methods cannot work, CsiNet retains effective beamforming gain.

Subfields: Aerospace Engineering, Electrical and Electronic Engineering

View Paper: https://www.semanticscholar.org/paper/391596b543afaf45f5d0213719b4863596c1dd11

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Electrical and Electronic Engineering

Citations in the corpus

Deep Learning-Based CSI Feedback Approach for Time-Varying Massive MIMO Channels (2018, methodology): “The CsiNet in [8] demonstrates remarkable performance in CSI sensing and reconstruction.”
Massive MIMO Detection Techniques: A Survey (2019, methodology): “learning in communication system such as the detector design for the MIMO transmission [148], [149], [150], [151], the channel estimation based massive MIMO [152], [153], and the pilot allocation for the massive MIMO [154].”
DeepMIMO: A Generic Deep Learning Dataset for Millimeter Wave and Massive MIMO Applications (2019, methodology): “While some methodologies for MIMO data generation have been proposed before for mobile applications [17], [18], there is no available dataset for mmWave/massive MIMO that satis-ﬁes the mentioned requirements, to the best of our knowledge.”
Model-Driven Deep Learning for MIMO Detection (2020, background): “Recently, it has been applied in physical layer communications [11]–[13], such as channel estimation [14]–[16], CSI feedback [17], signal detection [18]–[26], channel coding [27], [28], and end-to-end transceiver design [29], [30].”
Model-Driven Deep Learning for Physical Layer Communications (2018, background): “4, a DL-based network, called CsiNet, has been proposed in [18] to reduce the feedback overhead in massive MIMO systems.”
Massive Access for 5G and Beyond (2020, methodology): “Moreover, deep learning was adopted to compress the sparse CSI in [106], such that the amount of feedback becomes affordable.”
A Very Brief Introduction to Machine Learning With Applications to Communication Systems (2018, methodology): “For channels with a complex structure with unavailable channel models or with unknown optimal compression algorithms, autoencoders can be used to compress channel state information for the purpose of feedback on frequency-division duplex links [69].”
A Deep Learning Framework for Optimization of MISO Downlink Beamforming (2019, background): “A lot of research has attempted to use DL to address physical layer issues, including channel decoding [17, 18], detection [19–21], channel estimation [22–24], and resource management [25–32].”
Deep Learning in Physical Layer Communications (2018, methodology): “It has been shown that CsiNet remarkably outperforms traditional CS-based methods in terms of both compression ratio and recovery speed.”
Reconfigurable Intelligent Surfaces: Principles and Opportunities (2020, background): “However, CSI acquisition becomes more challenging due to the large number of antenna elements in massive MIMO systems [108].”
Deep Learning for Physical-Layer 5G Wireless Techniques: Opportunities, Challenges and Solutions (2019, background): “Other state-of-the-art work has provided interesting ideas of deep learning-based channel estimation [10].”
Wireless Image Transmission Using Deep Source Channel Coding With Attention Modules (2020, background): “The impressive performance gain brought by deep learning (DL) in computer vision [11] and natural language processing [12] in recent years have been widely concerned by researchers to deal with various wireless communication problems such as routing control [13], channel decoding [14], channel state”
Deep CNN-Based Channel Estimation for mmWave Massive MIMO Systems (2019, methodology): “To reduce the CSI feedback overhead of the frequency duplex division (FDD) massive MIMO system, DL has been employed in [17] to compress the channel into a low”
Convolutional Neural Network-Based Multiple-Rate Compressive Sensing for Massive MIMO CSI Feedback: Design, Simulation, and Analysis (2019, methodology): “In the CsiNet in [21], the encoder acts as the role of compression module instead of randomized Gaussian matrix at the UE and the decoder is regarded as the decompression module at the BS.”
Massive MIMO is a Reality - What is Next? Five Promising Research Directions for Antenna Arrays (2019, methodology): “Some recent examples include MIMO detection [141], blind channel equalization [142], channel estimation [143] and feedback [144], Massive MIMO power control [145], [146], user positioning [90], as well as full learning of transceiver implementations [147].”
Covert Communications: A Comprehensive Survey (2023, background): “However, they cannot stop the malicious eavesdroppers from improving their decoding techniques to acquire the transmitted contents, including the distributed computing to tackle the password or the advanced channel state information (CSI) estimation [20], [21].”