Deep Learning Techniques for Decoding Polar Codes

Gross, Warren J.; Doan, Nghia; Mambou, Elie Ngomseu; Hashemi, Seyyed Ali

doi:10.1002/9781119562306.ch15

Cited by 7 publications

(1 citation statement)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [ 30 ], a double long short-term memory (DLSTM) neural network was proposed to improve the performance of short block length by clipping frozen bits to enhance prediction accuracy. Deep learning-based decoders perform better than traditional decoders [ 31 ]. However, most decoding technologies treat deep neural networks as black boxes, where the decoding network consists of only a few fully connected or convolutional layers that take polar codes as inputs to the network and output decoded symbols directly.…”

Section: Introductionmentioning

confidence: 99%

DIR-Net: Deep Residual Polar Decoding Network Based on Information Refinement

Song

Feng

Wang

2022

Entropy

View full text Add to dashboard Cite

Polar codes are closer to the Shannon limit with lower complexity in coding and decoding. As traditional decoding techniques suffer from high latency and low throughput, with the development of deep learning technology, some deep learning-based decoding methods have been proposed to solve these problems. Usually, the deep neural network is treated as a black box and learns to map the polar codes with noise to the original information code directly. In fact, it is difficult for the network to distinguish between valid and interfering information, which leads to limited BER performance. In this paper, a deep residual network based on information refinement (DIR-NET) is proposed for decoding polar-coded short packets. The proposed method works to fully distinguish the effective and interference information in the codewords, thus obtaining a lower bit error rate. To achieve this goal, we design a two-stage decoding network, including a denoising subnetwork and decoding subnetwork. This structure can further improve the accuracy of the decoding method. Furthermore, we construct the whole network solely on the basis of the attention mechanism. It has a stronger information extraction ability than the traditional neural network structure. Benefiting from cascaded attention modules, information can be filtered and refined step-by-step, thus obtaining a low bit error rate. The simulation results show that DIR-Net outperforms existing decoding methods in terms of BER performance under both AWGN channels and flat fading channels.

show abstract

Section: Introductionmentioning

confidence: 99%