Reinforcement Learning for Channel Coding: Learned Bit-Flipping Decoding

@article{Carpi2019ReinforcementLF,
  title={Reinforcement Learning for Channel Coding: Learned Bit-Flipping Decoding},
  author={Fabrizio Carpi and Christian H{\"a}ger and Marco Martal{\`o} and Riccardo Raheli and Henry D. Pfister},
  journal={2019 57th Annual Allerton Conference on Communication, Control, and Computing (Allerton)},
  year={2019},
  pages={922-929}
}
In this paper, we use reinforcement learning to find effective decoding strategies for binary linear codes. We start by reviewing several iterative decoding algorithms that involve a decision-making process at each step, including bit-flipping (BF) decoding, residual belief propagation, and anchor decoding. We then illustrate how such algorithms can be mapped to Markov decision processes allowing for data-driven learning of optimal decision strategies, rather than basing decisions on heuristics… 
Reinforcement Learning for Bit-Flipping Decoding of Polar Codes
TLDR
A new algorithm combining reinforcement learning and SC flip (SCF) decoding of polar codes, which is called a Q-learning-assisted SCF (QLSCF), which uses reinforcement learning technology to select candidate bits for the SC flipping decoding.
A Reinforcement Learning Based Decoding Method of Short Polar Codes
  • Jian Gao, K. Niu
  • Computer Science
    2021 IEEE Wireless Communications and Networking Conference Workshops (WCNCW)
  • 2021
TLDR
In this paper, reinforcement learning is applied to find effective decoding strategies for short polar codes and an adaptive Q-table approach is proposed for data-driven learning of optimal decision strategies.
Belief Propagation Decoding of Short Graph-Based Channel Codes via Reinforcement Learning
TLDR
A novel graph-induced CN clustering approach to partition the state space of the MDP in such a way that dependencies between clusters are minimized and some of the proposed RL schemes not only improve the decoding performance, but also reduce the decoding complexity dramatically once the scheduling policy is learned.
RELDEC: Reinforcement Learning-Based Decoding of Moderate Length LDPC Codes
TLDR
The proposed RELDEC scheme significantly outperforms standard flooding and random sequential decoding for a variety of LDPC codes, including codes designed for 5G new radio.
Deepcode: Feedback Codes via Deep Learning
TLDR
This work presents the first family of codes obtained via deep learning, which significantly outperforms state-of-the-art codes designed over several decades of research, and demonstrates several desirable properties of the codes.
A Machine Learning Based Multi-flips Successive Cancellation Decoding Scheme of Polar Codes
TLDR
A machine learning based multi-flips SC decoding scheme (ML-MSCF), which can improve the performance of the SCF decoding algorithm with multiple flips based on the long short-term memory (LSTM) network and reinforcement learning (RL).
perm2vec: Attentive Graph Permutation Selection for Decoding of Error Correction Codes
TLDR
This work is the first to leverage the benefits of the neural Transformer networks in physical layer communication systems, and presents a data-driven framework for permutation selection, combining domain knowledge with machine learning concepts such as node embedding and self-attention.
Autoregressive Belief Propagation for Decoding Block Codes
TLDR
This work revisits recent methods that employ graph neural networks for decoding error correcting codes and employ messages that are computed in an autoregressive manner and obtains a bit error rate that outperforms the latest methods by a sizable margin.
Cyclically Equivariant Neural Decoders for Cyclic Codes
TLDR
This work proposes a novel neural decoder for cyclic codes by exploiting their cyclically invariant property, and imposes a shift invariant structure on the weights of the neuralDecoder so that any cyclic shift of inputs results in the same cyclicshift of outputs.
KO codes: inventing nonlinear encoding and decoding for reliable wireless communication via deep-learning
TLDR
KO codes are constructed, a computationaly efficient family of deep-learning driven (encoder, decoder) pairs that outperform the state-of-the-art reliability performance on the standardized AWGN channel and pave way for the discovery of a much richer class of hitherto unexplored nonlinear algebraic structures.
...
1
2
3
...

References

SHOWING 1-10 OF 32 REFERENCES
On deep learning-based channel decoding
TLDR
The metric normalized validation error (NVE) is introduced in order to further investigate the potential and limitations of deep learning-based decoding with respect to performance and complexity.
Learned Belief-Propagation Decoding with Simple Scaling and SNR Adaptation
TLDR
This work considers the weighted belief-propagation (WBP) decoder recently proposed by Nachmani et al. and investigates parameter adapter networks (PANs) that learn the relation between the signal-to-noise ratio and the WBP parameters.
Decoding Reed-Muller Codes Using Minimum- Weight Parity Checks
TLDR
The main idea is to apply iterative decoding to a highly-redundant parity-check (PC) matrix that contains only the minimum-weight dual codewords as rows, and proposes a method to tailor the PC matrix to the received observation by selecting only a small fraction of useful minimum-weights before decoding begins.
Communication Algorithms via Deep Learning
TLDR
It is shown that creatively designed and trained RNN architectures can decode well known sequential codes such as the convolutional and turbo codes with close to optimal performance on the additive white Gaussian noise (AWGN) channel.
Deep Learning for Decoding of Linear Codes - A Syndrome-Based Approach
TLDR
This work presents a novel framework for applying deep neural networks (DNN) to soft decoding of linear codes at arbitrary block lengths, and proves analytically that this approach does not involve any intrinsic performance penalty, and guarantees the generalization of performance obtained during training.
Soft-decision decoding of linear block codes based on ordered statistics
TLDR
A novel approach to soft decision decoding for binary linear block codes that achieves a desired error performance progressively in a number of stages and is terminated at the stage where either near-optimum error performance or a desired level of error performance is achieved.
Approaching Miscorrection-Free Performance of Product Codes With Anchor Decoding
TLDR
A novel iterative decoding algorithm for PCs which can detect and avoid most miscorrections, and can be used to decode many recently proposed classes of generalized PCs, such as staircase, braided, and half-product codes.
Soft-decision decoding of Reed-Muller codes: recursive lists
TLDR
Simulation results show that for all RM codes of length 256 and many subcodes of length 512, these algorithms approach maximum-likelihood (ML) performance within a margin of 0.1 dB.
Learning to Flip Successive Cancellation Decoding of Polar Codes with LSTM Networks
  • Xianbin Wang, Huazi Zhang, +4 authors J. Wang
  • Computer Science, Mathematics
    2019 IEEE 30th Annual International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC)
  • 2019
TLDR
Simulation results show that the proposed approach identifies error bits more accurately and achieves better block error rate performance than the state-of-the-art SC flip algorithms.
A decoding algorithm for finite-geometry LDPC codes
TLDR
A new low-complexity algorithm to decode low-density parity-check (LDPC) codes that achieves an appealing tradeoff between performance and complexity for FG-LDPC codes.
...
1
2
3
4
...