Neural Network Optimal UW-OFDM

Gergö Bognar; Stefan Baumgartner; Oliver Lang; Mario Huemer

doi:10.1109/IEEECONF53345.2021.9723117

Neural Network Optimal UW-OFDM

Gergö Bognar, Stefan Baumgartner, Oliver Lang, Mario Huemer

Institute of Signal Processing

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

Abstract

We propose a novel end-to-end learning scheme for wireless communication systems employing the unique word (UW)-OFDM signaling scheme. The work is motivated by the recent advances of machine learning in channel equalization and data estimation. Our idea is to design a non-systematically encoded UW-OFDM system optimal for neural network (NN) based estimators. To this order, we introduce model-based neural network architectures that optimize the transmitter and receiver sides, i.e. the UW-OFDM symbol generation and the NN data estimation together for minimal bit error ratio (BER). The proposed model is evaluated in a simulation environment, and compared with NN-based and traditional estimators.

Original language	English
Title of host publication	Proceedings of the Asilomar Conference on Signals, Systems, and Computers (ACSSC 2021)
Publisher	IEEE
Pages	389-394
Number of pages	6
ISBN (Print)	978-1-6654-5828-3
DOIs	https://doi.org/10.1109/IEEECONF53345.2021.9723117
Publication status	Published - Nov 2021

Fields of science

202040 Transmission technology
102019 Machine learning
202 Electrical Engineering, Electronics, Information Engineering
202015 Electronics
202022 Information technology
202030 Communication engineering
202037 Signal processing

JKU Focus areas

Digital Transformation

Access to Document

10.1109/IEEECONF53345.2021.9723117

Cite this

@inproceedings{708c74f383df48d3945aa19ebf8c8323,

title = "Neural Network Optimal UW-OFDM",

abstract = "We propose a novel end-to-end learning scheme for wireless communication systems employing the unique word (UW)-OFDM signaling scheme. The work is motivated by the recent advances of machine learning in channel equalization and data estimation. Our idea is to design a non-systematically encoded UW-OFDM system optimal for neural network (NN) based estimators. To this order, we introduce model-based neural network architectures that optimize the transmitter and receiver sides, i.e. the UW-OFDM symbol generation and the NN data estimation together for minimal bit error ratio (BER). The proposed model is evaluated in a simulation environment, and compared with NN-based and traditional estimators.",

author = "Gerg{\"o} Bognar and Stefan Baumgartner and Oliver Lang and Mario Huemer",

year = "2021",

month = nov,

doi = "10.1109/IEEECONF53345.2021.9723117",

language = "English",

isbn = "978-1-6654-5828-3",

pages = "389--394",

booktitle = "Proceedings of the Asilomar Conference on Signals, Systems, and Computers (ACSSC 2021)",

publisher = "IEEE",

}

TY - GEN

T1 - Neural Network Optimal UW-OFDM

AU - Bognar, Gergö

AU - Baumgartner, Stefan

AU - Lang, Oliver

AU - Huemer, Mario

PY - 2021/11

Y1 - 2021/11

N2 - We propose a novel end-to-end learning scheme for wireless communication systems employing the unique word (UW)-OFDM signaling scheme. The work is motivated by the recent advances of machine learning in channel equalization and data estimation. Our idea is to design a non-systematically encoded UW-OFDM system optimal for neural network (NN) based estimators. To this order, we introduce model-based neural network architectures that optimize the transmitter and receiver sides, i.e. the UW-OFDM symbol generation and the NN data estimation together for minimal bit error ratio (BER). The proposed model is evaluated in a simulation environment, and compared with NN-based and traditional estimators.

AB - We propose a novel end-to-end learning scheme for wireless communication systems employing the unique word (UW)-OFDM signaling scheme. The work is motivated by the recent advances of machine learning in channel equalization and data estimation. Our idea is to design a non-systematically encoded UW-OFDM system optimal for neural network (NN) based estimators. To this order, we introduce model-based neural network architectures that optimize the transmitter and receiver sides, i.e. the UW-OFDM symbol generation and the NN data estimation together for minimal bit error ratio (BER). The proposed model is evaluated in a simulation environment, and compared with NN-based and traditional estimators.

UR - https://ieeexplore.ieee.org/document/9723117

U2 - 10.1109/IEEECONF53345.2021.9723117

DO - 10.1109/IEEECONF53345.2021.9723117

M3 - Conference proceedings

SN - 978-1-6654-5828-3

SP - 389

EP - 394

BT - Proceedings of the Asilomar Conference on Signals, Systems, and Computers (ACSSC 2021)

PB - IEEE

ER -

Neural Network Optimal UW-OFDM

Abstract

Fields of science

JKU Focus areas

Access to Document

Other files and links

JKU LIT SAL eSPML Lab

Cite this

Neural Network Optimal UW-OFDM

Abstract

Fields of science

JKU Focus areas

Access to Document

Other files and links

Projects

JKU LIT SAL eSPML Lab

Cite this