On deep learning-based channel decoding

By Tobias Gruber, Sebastian Cammerer, Jakob Hoydis, Stephan ten Brink. Annual Conference on Information Sciences and Systems, 2017

555 citations recorded by the source corpus.

We revisit the idea of using deep neural networks for one-shot decoding of random and structured codes, such as polar codes. Although it is possible to achieve maximum a posteriori (MAP) bit error rate (BER) performance for both code families and for short codeword lengths, we observe that (i) structured codes are easier to learn and (ii) the neural network is able to generalize to codewords that it has never seen during training for structured, but not for random codes. These results provide some evidence that neural networks can learn a form of decoding algorithm, rather than only a simple classifier. We introduce the metric normalized validation error (NVE) in order to further investigate the potential and limitations of deep learning-based decoding with respect to performance and complexity.

Subfields: Artificial Intelligence, Computer Networks and Communications, Electrical and Electronic Engineering

View Paper: https://www.semanticscholar.org/paper/f3b944046602765fc1efc94264d291e4052a1a84

How researchers see this paper

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

In Artificial Intelligence

In Computer Networks and Communications

In Electrical and Electronic Engineering

Citations in the corpus

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”
White Paper on Broadband Connectivity in 6G (2020, background): “Furthermore, modiﬁed polar code constructions can be adopted to improve the performance of iterative algorithms, potentially with the aid of machine learning such as deep unfolding [98, 99].”
Deep Learning-Based Channel Estimation (2018, background): “…some approaches have been proposed to enhance the performance of different conventional algorithms including modulation recognition [7], channel encoding and decoding [8], signal detection [9], channel equalization [10], channel state information (CSI) feedback [11] and channel estimation [12].”
Grant-Free Non-Orthogonal Multiple Access for IoT: A Survey (2019, methodology): “For the physical layer, learning algorithms have been applied to demodulation [180], channel codes [181], BP for channel decoding [182], one-shot channel decoding [183], CS [184], coherent [185], and blind [186] detection.”
Robust Decentralized Joint Precoding using Team Deep Neural Network (2018, background): “For example, several groups have considered how to apply Machine Learning for decoding [14], [15], for detection [16], [17], for caching [18] and for resource allocation in interference channels [19].”
Deep Learning Based Communication Over the Air (2017, background): “Among them are belief propagation for channel decoding [16], [17], one-shot channel decoding [18], compressed sensing [19], [20], coherent [21] and blind [22] detection for multiple-input multiple-output (MIMO) systems, detection algorithms for molecular communications [23], learning of…”
A Very Brief Introduction to Machine Learning With Applications to Communication Systems (2018, methodology): “ received signals – the inputs to the decoding machine – and correct transmitted messages – the outputs of the decoding machine – as examples to be used for the training of a given black-box function [9]. The potential advantages of the machine learning approach over the conventional engineering des”
Deep learning for wireless physical layer: Opportunities and challenges (2017, methodology): “…to enhance certain parts of the conventional communication system (e.g., modulation recognition [1], channel encoding and decoding [25], [26], [27], [28], [29], [30], and channel estimation and detection [6], [4], [31], [32]) and to replace the total system with a novel architecture on the…”
Team Deep Neural Networks for Interference Channels (2018, methodology): “In [15] a framework to incorporate machine learning in cognitive radio is described while its application to coding is discussed in [16], [17].”
An Introduction to Deep Learning for the Physical Layer (2017, background): “Some of us have also demonstrated initial results in the area of learned end-to-end communications systems [30] as well as considered the problems of modulation recognition [31], signal compression [32], and channel decoding [33], [34] with state-of-the art DL tools.”
6G Wireless Systems: Vision, Requirements, Challenges, Insights, and Opportunities (2020, background): “New coding strategies should encompass both forward error correction and include novel iterative retransmission/feedback mechanisms [85] and ML-based methods [86].”
Machine Learning in the Air (2019, unknown): “A fully DNN-based channel decoder is considered in [33].”