Blind channel equalization using variational autoencoders

Avi Caciularu; David Burshtein

doi:10.1109/ICCW.2018.8403666

Blind channel equalization using variational autoencoders

Avi Caciularu, David Burshtein

Tel Aviv University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

40 Scopus citations

Abstract

A new maximum likelihood estimation approach for blind channel equalization, using variational autoencoders (VAEs), is introduced. Significant and consistent improvements in the error rate of the reconstructed symbols, compared to constant modulus equalizers, are demonstrated. In fact, for the channels that were examined, the performance of the new VAE blind channel equalizer was close to the performance of a nonblind adaptive linear minimum mean square error equalizer. The new equalization method enables a significantly lower latency channel acquisition compared to the constant modulus algorithm (CMA). The VAE uses a convolutional neural network with two layers and a very small number of free parameters. Although the computational complexity of the new equalizer is higher compared to CMA, it is still reasonable, and the number of free parameters to estimate is small.

Original language	English
Title of host publication	2018 IEEE International Conference on Communications Workshops, ICC Workshops 2018 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1-6
Number of pages	6
ISBN (Electronic)	9781538643280
DOIs	https://doi.org/10.1109/ICCW.2018.8403666
State	Published - 3 Jul 2018
Externally published	Yes
Event	2018 IEEE International Conference on Communications Workshops, ICC Workshops 2018 - Kansas City, United States Duration: 20 May 2018 → 24 May 2018

Publication series

Name	2018 IEEE International Conference on Communications Workshops, ICC Workshops 2018 - Proceedings

Conference

Conference	2018 IEEE International Conference on Communications Workshops, ICC Workshops 2018
Country/Territory	United States
City	Kansas City
Period	20/05/18 → 24/05/18

Bibliographical note

Publisher Copyright:
© 2018 IEEE.

Funding

This research was supported by the Israel Science Foundation, grant no. 1082/13. We would like to thank Sarvraj Singh Ranhotra for sharing with us the simulations code in [26].

Funders	Funder number
Israel Science Foundation	1082/13

Keywords

Blind channel equalization
Constant modulus algorithm
Convolutional neural networks
Deep learning
Maximum likelihood
Variational autoencoders

Access to Document

10.1109/ICCW.2018.8403666

Cite this

Caciularu, A., & Burshtein, D. (2018). Blind channel equalization using variational autoencoders. In 2018 IEEE International Conference on Communications Workshops, ICC Workshops 2018 - Proceedings (pp. 1-6). (2018 IEEE International Conference on Communications Workshops, ICC Workshops 2018 - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICCW.2018.8403666

@inproceedings{33c7bcd73b5c46e2846bcab43a74034d,

title = "Blind channel equalization using variational autoencoders",

abstract = "A new maximum likelihood estimation approach for blind channel equalization, using variational autoencoders (VAEs), is introduced. Significant and consistent improvements in the error rate of the reconstructed symbols, compared to constant modulus equalizers, are demonstrated. In fact, for the channels that were examined, the performance of the new VAE blind channel equalizer was close to the performance of a nonblind adaptive linear minimum mean square error equalizer. The new equalization method enables a significantly lower latency channel acquisition compared to the constant modulus algorithm (CMA). The VAE uses a convolutional neural network with two layers and a very small number of free parameters. Although the computational complexity of the new equalizer is higher compared to CMA, it is still reasonable, and the number of free parameters to estimate is small.",

keywords = "Blind channel equalization, Constant modulus algorithm, Convolutional neural networks, Deep learning, Maximum likelihood, Variational autoencoders",

author = "Avi Caciularu and David Burshtein",

note = "Publisher Copyright: {\textcopyright} 2018 IEEE.; 2018 IEEE International Conference on Communications Workshops, ICC Workshops 2018 ; Conference date: 20-05-2018 Through 24-05-2018",

year = "2018",

month = jul,

day = "3",

doi = "10.1109/ICCW.2018.8403666",

language = "אנגלית",

series = "2018 IEEE International Conference on Communications Workshops, ICC Workshops 2018 - Proceedings",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "1--6",

booktitle = "2018 IEEE International Conference on Communications Workshops, ICC Workshops 2018 - Proceedings",

address = "ארצות הברית",

}

Caciularu, A & Burshtein, D 2018, Blind channel equalization using variational autoencoders. in 2018 IEEE International Conference on Communications Workshops, ICC Workshops 2018 - Proceedings. 2018 IEEE International Conference on Communications Workshops, ICC Workshops 2018 - Proceedings, Institute of Electrical and Electronics Engineers Inc., pp. 1-6, 2018 IEEE International Conference on Communications Workshops, ICC Workshops 2018, Kansas City, United States, 20/05/18. https://doi.org/10.1109/ICCW.2018.8403666

Blind channel equalization using variational autoencoders. / Caciularu, Avi; Burshtein, David.
2018 IEEE International Conference on Communications Workshops, ICC Workshops 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2018. p. 1-6 (2018 IEEE International Conference on Communications Workshops, ICC Workshops 2018 - Proceedings).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Blind channel equalization using variational autoencoders

AU - Caciularu, Avi

AU - Burshtein, David

PY - 2018/7/3

Y1 - 2018/7/3

N2 - A new maximum likelihood estimation approach for blind channel equalization, using variational autoencoders (VAEs), is introduced. Significant and consistent improvements in the error rate of the reconstructed symbols, compared to constant modulus equalizers, are demonstrated. In fact, for the channels that were examined, the performance of the new VAE blind channel equalizer was close to the performance of a nonblind adaptive linear minimum mean square error equalizer. The new equalization method enables a significantly lower latency channel acquisition compared to the constant modulus algorithm (CMA). The VAE uses a convolutional neural network with two layers and a very small number of free parameters. Although the computational complexity of the new equalizer is higher compared to CMA, it is still reasonable, and the number of free parameters to estimate is small.

AB - A new maximum likelihood estimation approach for blind channel equalization, using variational autoencoders (VAEs), is introduced. Significant and consistent improvements in the error rate of the reconstructed symbols, compared to constant modulus equalizers, are demonstrated. In fact, for the channels that were examined, the performance of the new VAE blind channel equalizer was close to the performance of a nonblind adaptive linear minimum mean square error equalizer. The new equalization method enables a significantly lower latency channel acquisition compared to the constant modulus algorithm (CMA). The VAE uses a convolutional neural network with two layers and a very small number of free parameters. Although the computational complexity of the new equalizer is higher compared to CMA, it is still reasonable, and the number of free parameters to estimate is small.

KW - Blind channel equalization

KW - Constant modulus algorithm

KW - Convolutional neural networks

KW - Deep learning

KW - Maximum likelihood

KW - Variational autoencoders

UR - http://www.scopus.com/inward/record.url?scp=85050284755&partnerID=8YFLogxK

U2 - 10.1109/ICCW.2018.8403666

DO - 10.1109/ICCW.2018.8403666

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:85050284755

T3 - 2018 IEEE International Conference on Communications Workshops, ICC Workshops 2018 - Proceedings

SP - 1

EP - 6

BT - 2018 IEEE International Conference on Communications Workshops, ICC Workshops 2018 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2018 IEEE International Conference on Communications Workshops, ICC Workshops 2018

Y2 - 20 May 2018 through 24 May 2018

ER -

Caciularu A, Burshtein D. Blind channel equalization using variational autoencoders. In 2018 IEEE International Conference on Communications Workshops, ICC Workshops 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2018. p. 1-6. (2018 IEEE International Conference on Communications Workshops, ICC Workshops 2018 - Proceedings). doi: 10.1109/ICCW.2018.8403666

Blind channel equalization using variational autoencoders

Abstract

Publication series

Conference

Bibliographical note

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this