Neural Network Optimal UW-OFDM

Gergő Borgnár; Stefan Baumgartner; Oliver Lang; Mario Huemer

doi:10.1109/IEEECONF53345.2021.9723117

Neural Network Optimal UW-OFDM

Gergő Borgnár, Stefan Baumgartner, Oliver Lang, Mario Huemer

Research output: Conference proceeding/Chapter in Book/Report/ › Conference Paper › 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	2021 55th Asilomar Conference on Signals, Systems, and Computers
Pages	389-394
Number of pages	6
DOIs	https://doi.org/10.1109/IEEECONF53345.2021.9723117
Publication status	Published - 3 Nov 2021
Externally published	Yes
Event	2021 55th Asilomar Conference on Signals, Systems, and Computers - Pacific Grove, CA, USA Duration: 31 Oct 2021 → 3 Nov 2021

Conference

Conference	2021 55th Asilomar Conference on Signals, Systems, and Computers
Period	31/10/21 → 3/11/21

Keywords

Wireless communication
Computational modeling
Estimation
Artificial neural networks
Computer architecture
Generators
Data models

Access to Document

10.1109/IEEECONF53345.2021.9723117

https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9723117

Cite this

@inproceedings{8366c54b4b004606843dc27c2eef7f3e,

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.",

keywords = "Wireless communication, Computational modeling, Estimation, Artificial neural networks, Computer architecture, Generators, Data models",

author = "Gerg{\H o} Borgn{\'a}r and Stefan Baumgartner and Oliver Lang and Mario Huemer",

year = "2021",

month = nov,

day = "3",

doi = "10.1109/IEEECONF53345.2021.9723117",

language = "English",

isbn = "978-1-6654-5829-0",

pages = "389--394",

booktitle = "2021 55th Asilomar Conference on Signals, Systems, and Computers",

note = "2021 55th Asilomar Conference on Signals, Systems, and Computers ; Conference date: 31-10-2021 Through 03-11-2021",

}

TY - GEN

T1 - Neural Network Optimal UW-OFDM

AU - Borgnár, Gergő

AU - Baumgartner, Stefan

AU - Lang, Oliver

AU - Huemer, Mario

PY - 2021/11/3

Y1 - 2021/11/3

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.

KW - Wireless communication

KW - Computational modeling

KW - Estimation

KW - Artificial neural networks

KW - Computer architecture

KW - Generators

KW - Data models

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

U2 - 10.1109/IEEECONF53345.2021.9723117

DO - 10.1109/IEEECONF53345.2021.9723117

M3 - Conference Paper

SN - 978-1-6654-5829-0

SP - 389

EP - 394

BT - 2021 55th Asilomar Conference on Signals, Systems, and Computers

T2 - 2021 55th Asilomar Conference on Signals, Systems, and Computers

Y2 - 31 October 2021 through 3 November 2021

ER -

Neural Network Optimal UW-OFDM

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this