IEEE Transactions on Wireless Communications, volume 20, issue 6, pages 3925-3940

DeepRx: Fully Convolutional Deep Learning Receiver

Mikko Honkala 1
Dani Korpi 1
J. M. J. Huttunen 1
Publication typeJournal Article
Publication date2021-06-01
Quartile SCImago
Q1
Quartile WOS
Q1
Impact factor10.4
ISSN15361276, 15582248
Computer Science Applications
Electrical and Electronic Engineering
Applied Mathematics
Abstract
Deep learning has solved many problems that are out of reach of heuristic algorithms. It has also been successfully applied in wireless communications, even though the current radio systems are well-understood and optimal algorithms exist for many tasks. While some gains have been obtained by learning individual parts of a receiver, a better approach is to jointly learn the whole receiver. This, however, often results in a challenging nonlinear problem, for which the optimal solution is infeasible to implement. To this end, we propose a deep fully convolutional neural network, DeepRx, which executes the whole receiver pipeline from frequency domain signal stream to uncoded bits in a 5G-compliant fashion. We facilitate accurate channel estimation by constructing the input of the convolutional neural network in a very specific manner using both the data and pilot symbols. Also, DeepRx outputs soft bits that are compatible with the channel coding used in 5G systems. Using 3GPP-defined channel models, we demonstrate that DeepRx outperforms traditional methods. We also show that the high performance can likely be attributed to DeepRx learning to utilize the known constellation points of the unknown data symbols, together with the local symbol distribution, for improved detection accuracy.

Top-30

Citations by journals

2
4
6
8
10
12
14
IEEE Transactions on Wireless Communications
13 publications, 15.12%
IEEE Journal on Selected Areas in Communications
3 publications, 3.49%
IEEE Communications Letters
3 publications, 3.49%
IEEE Transactions on Communications
3 publications, 3.49%
IEEE Transactions on Cognitive Communications and Networking
2 publications, 2.33%
Signals and Communication Technology
2 publications, 2.33%
IEEE Wireless Communications
2 publications, 2.33%
Applied Sciences (Switzerland)
1 publication, 1.16%
Wireless Personal Communications
1 publication, 1.16%
Journal of Physics: Conference Series
1 publication, 1.16%
Lecture Notes in Electrical Engineering
1 publication, 1.16%
ETRI Journal
1 publication, 1.16%
INSIGHT
1 publication, 1.16%
Mobile Networks and Applications
1 publication, 1.16%
Lecture Notes in Networks and Systems
1 publication, 1.16%
IEEE Access
1 publication, 1.16%
IEEE Open Journal of the Communications Society
1 publication, 1.16%
IEEE BITS the Information Theory Magazine
1 publication, 1.16%
Future Internet
1 publication, 1.16%
IEEE Transactions on Circuits and Systems I: Regular Papers
1 publication, 1.16%
IEEE Wireless Communications Letters
1 publication, 1.16%
Signal Processing
1 publication, 1.16%
IEEE Transactions on Vehicular Technology
1 publication, 1.16%
Mathematics
1 publication, 1.16%
2
4
6
8
10
12
14

Citations by publishers

10
20
30
40
50
60
70
80
IEEE
72 publications, 83.72%
Springer Nature
6 publications, 6.98%
Multidisciplinary Digital Publishing Institute (MDPI)
3 publications, 3.49%
Wiley
2 publications, 2.33%
IOP Publishing
1 publication, 1.16%
Russian State Vocational Pedagogical University
1 publication, 1.16%
Elsevier
1 publication, 1.16%
10
20
30
40
50
60
70
80
  • We do not take into account publications without a DOI.
  • Statistics recalculated only for publications connected to researchers, organizations and labs registered on the platform.
  • Statistics recalculated weekly.

Are you a researcher?

Create a profile to get free access to personal recommendations for colleagues and new articles.
Metrics
Share
Cite this
GOST |
Cite this
GOST Copy
Honkala M. et al. DeepRx: Fully Convolutional Deep Learning Receiver // IEEE Transactions on Wireless Communications. 2021. Vol. 20. No. 6. pp. 3925-3940.
GOST all authors (up to 50) Copy
Honkala M., Korpi D., Huttunen J. M. J. DeepRx: Fully Convolutional Deep Learning Receiver // IEEE Transactions on Wireless Communications. 2021. Vol. 20. No. 6. pp. 3925-3940.
RIS |
Cite this
RIS Copy
TY - JOUR
DO - 10.1109/twc.2021.3054520
UR - https://doi.org/10.1109/twc.2021.3054520
TI - DeepRx: Fully Convolutional Deep Learning Receiver
T2 - IEEE Transactions on Wireless Communications
AU - Honkala, Mikko
AU - Korpi, Dani
AU - Huttunen, J. M. J.
PY - 2021
DA - 2021/06/01
PB - IEEE
SP - 3925-3940
IS - 6
VL - 20
SN - 1536-1276
SN - 1558-2248
ER -
BibTex |
Cite this
BibTex Copy
@article{2021_Honkala,
author = {Mikko Honkala and Dani Korpi and J. M. J. Huttunen},
title = {DeepRx: Fully Convolutional Deep Learning Receiver},
journal = {IEEE Transactions on Wireless Communications},
year = {2021},
volume = {20},
publisher = {IEEE},
month = {jun},
url = {https://doi.org/10.1109/twc.2021.3054520},
number = {6},
pages = {3925--3940},
doi = {10.1109/twc.2021.3054520}
}
MLA
Cite this
MLA Copy
Honkala, Mikko, et al. “DeepRx: Fully Convolutional Deep Learning Receiver.” IEEE Transactions on Wireless Communications, vol. 20, no. 6, Jun. 2021, pp. 3925-3940. https://doi.org/10.1109/twc.2021.3054520.
Found error?