Deep Learning Methods for Improved Decoding of Linear Codes

Abstract

The problem of low complexity, close to optimal, channel decoding of linear codes with short to moderate block length is considered. It is shown that deep learning methods can be used to improve a standard belief propagation decoder, despite the large example space. Similar improvements are obtained for the min-sum algorithm. It is also shown that tying the parameters of the decoders across iterations, so as to form a recurrent neural network architecture, can be implemented with comparable results. The advantage is that significantly less parameters are required. We also introduce a recurrent neural decoder architecture based on the method of successive relaxation. Improvements over standard belief propagation are also observed on sparser Tanner graph representations of the codes. Furthermore, we demonstrate that the neural belief propagation decoder can be used to improve the performance, or alternatively reduce the computational complexity, of a close to optimal decoder of short BCH codes.

Accepted To IEEE Journal Of Selected Topics In Signal Processing

References in corpus (7)

• Quantized Neural Networks: Training Neural Networks with Low Precision Weights and Activations
• Detection Algorithms for Communication Systems Using Deep Learning
• Deep Neural Network Probabilistic Decoder for Stabilizer Codes
• Code Design for Short Blocks: A Survey
• On Deep Learning-Based Channel Decoding
• RNN Decoding of Linear Block Codes
• Performance Comparison of Short-Length Error-Correcting Codes

Cited by in corpus (88)

• Deep Joint Source-Channel Coding for Wireless Image Transmission
• Learning to Detect
• Model-Driven Deep Learning for MIMO Detection
• Neural Network Detection of Data Sequences in Communication Systems
• White Paper on Broadband Connectivity in 6G
• DeepReceiver: A Deep Learning-Based Intelligent Receiver for Wireless Communications in the Physical Layer
• Scoring the Terabit/s Goal:Broadband Connectivity in 6G
• Artificial Neural Networks-Based Machine Learning for Wireless Networks: A Tutorial
• Machine Learning for Wireless Link Quality Estimation: A Survey
• Deep Learning Based Sphere Decoding
• Deep Learning-based Implicit CSI Feedback in Massive MIMO
• DNN-Aided Block Sparse Bayesian Learning for User Activity Detection and Channel Estimation in Grant-Free Non-Orthogonal Random Access
• Learned Belief-Propagation Decoding with Simple Scaling and SNR Adaptation
• Deep Learning Based Packet Detection and Carrier Frequency Offset Estimation in IEEE 802.11ah
• Learning to Continuously Optimize Wireless Resource in a Dynamic Environment: A Bilevel Optimization Perspective
• Deep Learning for Wireless Physical Layer: Opportunities and Challenges
• Exploiting Degeneracy in Belief Propagation Decoding of Quantum Codes
• A Survey of Multi-Access Edge Computing in 5G and Beyond: Fundamentals, Technology Integration, and State-of-the-Art
• Recent Advances in Deep Learning for Channel Coding: A Survey
• Learning to Continuously Optimize Wireless Resource In Episodically Dynamic Environment
• Towards Optimal Power Control via Ensembling Deep Neural Networks
• Syndrome-Enabled Unsupervised Learning for Neural Network-Based Polar Decoder and Jointly Optimized Blind Equalizer
• Decoding Short LDPC Codes via BP-RNN Diversity and Reliability-Based Post-Processing
• Cascade-Net: a New Deep Learning Architecture for OFDM Detection
• Erasure Decoding for Quantum LDPC Codes via Belief Propagation with Guided Decimation
• DeepTurbo: Deep Turbo Decoder
• LightCode: Light Analytical and Neural Codes for Channels with Feedback
• CMDNet: Learning a Probabilistic Relaxation of Discrete Variables for Soft Detection with Low Complexity
• Generalization Bounds for Neural Belief Propagation Decoders
• Deep Learning-Based Detection for Marker Codes over Insertion and Deletion Channels
• Low-complexity Near-optimum Symbol Detection Based on Neural Enhancement of Factor Graphs
• DeepSense: Enabling Carrier Sense in Low-Power Wide Area Networks Using Deep Learning
• A Multi-Objective Optimization Framework for URLLC with Decoding Complexity Constraints
• perm2vec: Graph Permutation Selection for Decoding of Error Correction Codes using Self-Attention
• DeepJSCC-f: Deep Joint Source-Channel Coding of Images with Feedback
• Machine Learning for Error Correction with Natural Redundancy
• Fixed-Symbol Aided Random Access Scheme for Machine-to-Machine Communications
• Deep Learning based Blind Symbol Packing Ratio Estimation for Faster-than-Nyquist Signaling
• Deep Learning Detection Networks in MIMO Decode-Forward Relay Channels
• Quaternary Neural Belief Propagation Decoding of Quantum LDPC Codes with Overcomplete Check Matrices
• Deep Ensemble of Weighted Viterbi Decoders for Tail-Biting Convolutional Codes
• A Very Brief Introduction to Machine Learning With Applications to Communication Systems
• Learning to Decode Protograph LDPC Codes
• Discriminative Mutual Information Estimators for Channel Capacity Learning
• Joint Neural Network Equalizer and Decoder
• Learned Decimation for Neural Belief Propagation Decoders
• Improved Syndrome-based Neural Decoder for Linear Block Codes
• Low-complexity Recurrent Neural Network-based Polar Decoder with Weight Quantization Mechanism
• On the Robustness of Deep Learning-aided Symbol Detectors to Varying Conditions and Imperfect Channel Knowledge
• TurboNet: A Model-driven DNN Decoder Based on Max-Log-MAP Algorithm for Turbo Code
• Convolutional Neural Network-aided Bit-flipping for Belief Propagation Decoding of Polar Codes
• ADMM-based Decoder for Binary Linear Codes Aided by Deep Learning
• Hybrid HMM Decoder For Convolutional Codes By Joint Trellis-Like Structure and Channel Prior
• Decision Directed Channel Estimation Based on Deep Neural Network k-step Predictor for MIMO Communications in 5G
• Autoregressive Belief Propagation for Decoding Block Codes
• Transformer-Based Decoding in Concatenated Coding Schemes Under Synchronization Errors
• A Neural Network Lattice Decoding Algorithm
• Machine Learning for Geometrically-Consistent Angular Spread Function Estimation in Massive MIMO
• A Novel Deep Neural Network Based Approach for Sparse Code Multiple Access
• A Deep-learning-based Joint Inference for Secure Spatial Modulation Receiver
• Deep Learning-Based Decoding of Constrained Sequence Codes
• Design of Capacity-Approaching Low-Density Parity-Check Codes using Recurrent Neural Networks
• Neural-Network-Optimized Degree-Specific Weights for LDPC MinSum Decoding
• Neural Network-based Equalizer by Utilizing Coding Gain in Advance
• Deep Learning for Joint MIMO Detection and Channel Decoding
• Efficient Active Deep Decoding of Linear Codes using Importance Sampling
• Data-Rate Driven Transmission Strategy for Deep Learning Based Communication Systems
• A Novel Sum-Product Detection Algorithm for Faster-than-Nyquist Signaling: A Deep Learning Approach
• DeepIC: Coding for Interference Channels via Deep Learning
• On the Design and Performance of Machine Learning Based Error Correcting Decoders
• Unsupervised Learning for Neural Network-based Polar Decoder via Syndrome Loss
• Iterative Error Decimation for Syndrome-Based Neural Network Decoders
• Modified EP MIMO Detection Algorithm with Deep Learning Parameters Selection
• Learning to Decode: Reinforcement Learning for Decoding of Sparse Graph-Based Channel Codes
• Self-dual codes over $GF(q)$ with symmetric generator matrices
• A lattice-based approach to the expressivity of deep ReLU neural networks
• Learning from the Syndrome
• Receiver Design for Faster-than-Nyquist Signaling: Deep-learning-based Architectures
• Polar Decoding on Sparse Graphs with Deep Learning
• Neural Network-Aided BCJR Algorithm for Joint Symbol Detection and Channel Decoding
• A Generalized Data Representation and Training-Performance Analysis for Deep Learning-Based Communications Systems
• Learning Physical-Layer Communication with Quantized Feedback
• What Can Machine Learning Teach Us about Communications?
• Learning Modulation Design for SWIPT with Nonlinear Energy Harvester: Large and Small Signal Power Regimes
• ProductAE: Towards Training Larger Channel Codes based on Neural Product Codes
• SNR-adaptive deep joint source-channel coding for wireless image transmission
• Model-Driven DNN Decoder for Turbo Codes: Design, Simulation and Experimental Results
• Deep Learning Assisted Sum-Product Detection Algorithm for Faster-than-Nyquist Signaling