Recurrent neural network based turbo decoding algorithms for different code rates

Application of deep learning to error control coding is gaining special attention and neural network architectures on decoding are approached to compare with conventional ones. Turbo codes conventionally use BCJR algorithm for decoding. In this paper, performances of neural Turbo decoder and deep le...

Full description

Saved in:
Bibliographic Details
Main Authors: Shridhar B. Devamane, Rajeshwari L. Itagi
Format: Article
Language:English
Published: Springer 2022-06-01
Series:Journal of King Saud University: Computer and Information Sciences
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S1319157820303323
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849324901251416064
author Shridhar B. Devamane
Rajeshwari L. Itagi
author_facet Shridhar B. Devamane
Rajeshwari L. Itagi
author_sort Shridhar B. Devamane
collection DOAJ
description Application of deep learning to error control coding is gaining special attention and neural network architectures on decoding are approached to compare with conventional ones. Turbo codes conventionally use BCJR algorithm for decoding. In this paper, performances of neural Turbo decoder and deep learning-based Turbo decoder are examined. A category of sequential codes are utilized to construct the RSC (Recursive Systematic Convolutional) codes as basic elements for Turbo encoder. Sequential codes suit the requirement of memory element present in convolution codes, which act as components for Turbo encoder. Turbo decoders are constructed by two means; as neural Turbo decoder and deep learning Turbo decoder. Both structures are based on recurrent neural network (RNN) architectures. RNN architectures are preferred due to the presence of memory as a feature. BER performance of both is compared with that of a convolutional Viterbi decoder in awgn channel. Both the structures are studied for different input data-lengths and code rates.
format Article
id doaj-art-e827307f738947eb8eaf73727a01bac6
institution Kabale University
issn 1319-1578
language English
publishDate 2022-06-01
publisher Springer
record_format Article
series Journal of King Saud University: Computer and Information Sciences
spelling doaj-art-e827307f738947eb8eaf73727a01bac62025-08-20T03:48:35ZengSpringerJournal of King Saud University: Computer and Information Sciences1319-15782022-06-013462666267910.1016/j.jksuci.2020.03.012Recurrent neural network based turbo decoding algorithms for different code ratesShridhar B. Devamane0Rajeshwari L. Itagi1Department of Electronics & Communication Engineering, APS College of Engineering, Bengaluru-560082, Karnataka, India; Corresponding author.Department of Electronics & Communication Engineering, KLE Institute of Technology, Hubballi-580030, Karnataka, IndiaApplication of deep learning to error control coding is gaining special attention and neural network architectures on decoding are approached to compare with conventional ones. Turbo codes conventionally use BCJR algorithm for decoding. In this paper, performances of neural Turbo decoder and deep learning-based Turbo decoder are examined. A category of sequential codes are utilized to construct the RSC (Recursive Systematic Convolutional) codes as basic elements for Turbo encoder. Sequential codes suit the requirement of memory element present in convolution codes, which act as components for Turbo encoder. Turbo decoders are constructed by two means; as neural Turbo decoder and deep learning Turbo decoder. Both structures are based on recurrent neural network (RNN) architectures. RNN architectures are preferred due to the presence of memory as a feature. BER performance of both is compared with that of a convolutional Viterbi decoder in awgn channel. Both the structures are studied for different input data-lengths and code rates.http://www.sciencedirect.com/science/article/pii/S1319157820303323Block error rateSignal to noise ratioAdditive white Gaussian noiseRecurrent neural networkDeep learningTurbo code
spellingShingle Shridhar B. Devamane
Rajeshwari L. Itagi
Recurrent neural network based turbo decoding algorithms for different code rates
Journal of King Saud University: Computer and Information Sciences
Block error rate
Signal to noise ratio
Additive white Gaussian noise
Recurrent neural network
Deep learning
Turbo code
title Recurrent neural network based turbo decoding algorithms for different code rates
title_full Recurrent neural network based turbo decoding algorithms for different code rates
title_fullStr Recurrent neural network based turbo decoding algorithms for different code rates
title_full_unstemmed Recurrent neural network based turbo decoding algorithms for different code rates
title_short Recurrent neural network based turbo decoding algorithms for different code rates
title_sort recurrent neural network based turbo decoding algorithms for different code rates
topic Block error rate
Signal to noise ratio
Additive white Gaussian noise
Recurrent neural network
Deep learning
Turbo code
url http://www.sciencedirect.com/science/article/pii/S1319157820303323
work_keys_str_mv AT shridharbdevamane recurrentneuralnetworkbasedturbodecodingalgorithmsfordifferentcoderates
AT rajeshwarilitagi recurrentneuralnetworkbasedturbodecodingalgorithmsfordifferentcoderates