DEEPTURBO: Deep Turbo Decoder

Yihan Jiang, Sreeram Kannan, Hyeji Kim, Sewoong Oh, Himanshu Asnani, Pramod Viswanath

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Present-day communication systems routinely use codes that approach the channel capacity when coupled with a computationally efficient decoder. However, the decoder is typically designed for the Gaussian noise channel, and is known to be sub-optimal for non-Gaussian noise distribution. Deep learning methods offer a new approach for designing decoders that can be trained and tailored for arbitrary channel statistics. We focus on Turbo codes, and propose (DEEPTURBO), a novel deep learning based architecture for Turbo decoding. The standard Turbo decoder (TURBO) iteratively applies the Bahl-Cocke-Jelinek-Raviv (BCJR) algorithm with an interleaver in the middle. A neural architecture for TURBO decoding, termed (NEURALBCJR), was proposed recently to create a module that imitates the BCJR algorithm using supervised learning, and to use the interleaver architecture along with this module, which is then fine-tuned using end-to-end training. However, knowledge of the BCJR algorithm is required to design such an architecture, which also constrains the resulting learnt decoder. Here we remedy this requirement and propose a fully end-to-end trained neural decoder-Deep Turbo Decoder (DEEPTURBO). With novel learnable decoder structure and training methodology, DEEPTURBO reveals superior performance under both AWGN and non-AWGN settings as compared to the other two decoders-TURBOand NEURALBCJR.

Original languageEnglish (US)
Title of host publication2019 IEEE 20th International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781538665282
DOIs
StatePublished - Jul 2019
Event20th IEEE International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2019 - Cannes, France
Duration: Jul 2 2019Jul 5 2019

Publication series

NameIEEE Workshop on Signal Processing Advances in Wireless Communications, SPAWC
Volume2019-July

Conference

Conference20th IEEE International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2019
CountryFrance
CityCannes
Period7/2/197/5/19

Fingerprint

Decoding
Turbo codes
Channel capacity
Supervised learning
Communication systems
Statistics
Deep learning

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Computer Science Applications
  • Information Systems

Cite this

Jiang, Y., Kannan, S., Kim, H., Oh, S., Asnani, H., & Viswanath, P. (2019). DEEPTURBO: Deep Turbo Decoder. In 2019 IEEE 20th International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2019 [8815400] (IEEE Workshop on Signal Processing Advances in Wireless Communications, SPAWC; Vol. 2019-July). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/SPAWC.2019.8815400

DEEPTURBO : Deep Turbo Decoder. / Jiang, Yihan; Kannan, Sreeram; Kim, Hyeji; Oh, Sewoong; Asnani, Himanshu; Viswanath, Pramod.

2019 IEEE 20th International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2019. Institute of Electrical and Electronics Engineers Inc., 2019. 8815400 (IEEE Workshop on Signal Processing Advances in Wireless Communications, SPAWC; Vol. 2019-July).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Jiang, Y, Kannan, S, Kim, H, Oh, S, Asnani, H & Viswanath, P 2019, DEEPTURBO: Deep Turbo Decoder. in 2019 IEEE 20th International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2019., 8815400, IEEE Workshop on Signal Processing Advances in Wireless Communications, SPAWC, vol. 2019-July, Institute of Electrical and Electronics Engineers Inc., 20th IEEE International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2019, Cannes, France, 7/2/19. https://doi.org/10.1109/SPAWC.2019.8815400
Jiang Y, Kannan S, Kim H, Oh S, Asnani H, Viswanath P. DEEPTURBO: Deep Turbo Decoder. In 2019 IEEE 20th International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2019. Institute of Electrical and Electronics Engineers Inc. 2019. 8815400. (IEEE Workshop on Signal Processing Advances in Wireless Communications, SPAWC). https://doi.org/10.1109/SPAWC.2019.8815400
Jiang, Yihan ; Kannan, Sreeram ; Kim, Hyeji ; Oh, Sewoong ; Asnani, Himanshu ; Viswanath, Pramod. / DEEPTURBO : Deep Turbo Decoder. 2019 IEEE 20th International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2019. Institute of Electrical and Electronics Engineers Inc., 2019. (IEEE Workshop on Signal Processing Advances in Wireless Communications, SPAWC).
@inproceedings{f4c4359ceee14011b76cec03c2998b03,
title = "DEEPTURBO: Deep Turbo Decoder",
abstract = "Present-day communication systems routinely use codes that approach the channel capacity when coupled with a computationally efficient decoder. However, the decoder is typically designed for the Gaussian noise channel, and is known to be sub-optimal for non-Gaussian noise distribution. Deep learning methods offer a new approach for designing decoders that can be trained and tailored for arbitrary channel statistics. We focus on Turbo codes, and propose (DEEPTURBO), a novel deep learning based architecture for Turbo decoding. The standard Turbo decoder (TURBO) iteratively applies the Bahl-Cocke-Jelinek-Raviv (BCJR) algorithm with an interleaver in the middle. A neural architecture for TURBO decoding, termed (NEURALBCJR), was proposed recently to create a module that imitates the BCJR algorithm using supervised learning, and to use the interleaver architecture along with this module, which is then fine-tuned using end-to-end training. However, knowledge of the BCJR algorithm is required to design such an architecture, which also constrains the resulting learnt decoder. Here we remedy this requirement and propose a fully end-to-end trained neural decoder-Deep Turbo Decoder (DEEPTURBO). With novel learnable decoder structure and training methodology, DEEPTURBO reveals superior performance under both AWGN and non-AWGN settings as compared to the other two decoders-TURBOand NEURALBCJR.",
author = "Yihan Jiang and Sreeram Kannan and Hyeji Kim and Sewoong Oh and Himanshu Asnani and Pramod Viswanath",
year = "2019",
month = "7",
doi = "10.1109/SPAWC.2019.8815400",
language = "English (US)",
series = "IEEE Workshop on Signal Processing Advances in Wireless Communications, SPAWC",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
booktitle = "2019 IEEE 20th International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2019",
address = "United States",

}

TY - GEN

T1 - DEEPTURBO

T2 - Deep Turbo Decoder

AU - Jiang, Yihan

AU - Kannan, Sreeram

AU - Kim, Hyeji

AU - Oh, Sewoong

AU - Asnani, Himanshu

AU - Viswanath, Pramod

PY - 2019/7

Y1 - 2019/7

N2 - Present-day communication systems routinely use codes that approach the channel capacity when coupled with a computationally efficient decoder. However, the decoder is typically designed for the Gaussian noise channel, and is known to be sub-optimal for non-Gaussian noise distribution. Deep learning methods offer a new approach for designing decoders that can be trained and tailored for arbitrary channel statistics. We focus on Turbo codes, and propose (DEEPTURBO), a novel deep learning based architecture for Turbo decoding. The standard Turbo decoder (TURBO) iteratively applies the Bahl-Cocke-Jelinek-Raviv (BCJR) algorithm with an interleaver in the middle. A neural architecture for TURBO decoding, termed (NEURALBCJR), was proposed recently to create a module that imitates the BCJR algorithm using supervised learning, and to use the interleaver architecture along with this module, which is then fine-tuned using end-to-end training. However, knowledge of the BCJR algorithm is required to design such an architecture, which also constrains the resulting learnt decoder. Here we remedy this requirement and propose a fully end-to-end trained neural decoder-Deep Turbo Decoder (DEEPTURBO). With novel learnable decoder structure and training methodology, DEEPTURBO reveals superior performance under both AWGN and non-AWGN settings as compared to the other two decoders-TURBOand NEURALBCJR.

AB - Present-day communication systems routinely use codes that approach the channel capacity when coupled with a computationally efficient decoder. However, the decoder is typically designed for the Gaussian noise channel, and is known to be sub-optimal for non-Gaussian noise distribution. Deep learning methods offer a new approach for designing decoders that can be trained and tailored for arbitrary channel statistics. We focus on Turbo codes, and propose (DEEPTURBO), a novel deep learning based architecture for Turbo decoding. The standard Turbo decoder (TURBO) iteratively applies the Bahl-Cocke-Jelinek-Raviv (BCJR) algorithm with an interleaver in the middle. A neural architecture for TURBO decoding, termed (NEURALBCJR), was proposed recently to create a module that imitates the BCJR algorithm using supervised learning, and to use the interleaver architecture along with this module, which is then fine-tuned using end-to-end training. However, knowledge of the BCJR algorithm is required to design such an architecture, which also constrains the resulting learnt decoder. Here we remedy this requirement and propose a fully end-to-end trained neural decoder-Deep Turbo Decoder (DEEPTURBO). With novel learnable decoder structure and training methodology, DEEPTURBO reveals superior performance under both AWGN and non-AWGN settings as compared to the other two decoders-TURBOand NEURALBCJR.

UR - http://www.scopus.com/inward/record.url?scp=85072340790&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85072340790&partnerID=8YFLogxK

U2 - 10.1109/SPAWC.2019.8815400

DO - 10.1109/SPAWC.2019.8815400

M3 - Conference contribution

AN - SCOPUS:85072340790

T3 - IEEE Workshop on Signal Processing Advances in Wireless Communications, SPAWC

BT - 2019 IEEE 20th International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2019

PB - Institute of Electrical and Electronics Engineers Inc.

ER -