End-to-end Physical Layer Communication using Bi-directional GRUs for ISI Channels

This paper presents a deep learning autoencoder for end-to-end physical-layer communications, in the presence of intersymbol interference (ISI) and additive white Gaussian noise (AWGN). Both the proposed transmitter and receiver employ the bi-directional gated recurrent unit (Bi-GRU) layers, and the...

Full description

Saved in:
Bibliographic Details
Published in2020 IEEE Globecom Workshops (GC Wkshps pp. 1 - 6
Main Authors Wu, Huihui, Zhang, Yitian, Zhao, Xueqing, Zhu, Ningbo, Coates, Mark
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.12.2020
Subjects
Bidirectional control
Deep learning
Demodulation
equalization
gated recurrent unit
Intersymbol interference
ISI channels
Logic gates
Maximum likelihood estimation
physical-layer autoencoders
Receivers
Transmitters
Online AccessGet full text

Cover

More Information
Summary:This paper presents a deep learning autoencoder for end-to-end physical-layer communications, in the presence of intersymbol interference (ISI) and additive white Gaussian noise (AWGN). Both the proposed transmitter and receiver employ the bi-directional gated recurrent unit (Bi-GRU) layers, and they are trained jointly with binary cross-entropy loss. By doing so, the transmitter learns tailored modulation constellations considering the channel impulse response while the receiver performs equalization and demodulation simultaneously. Experiments conducted over representative ISI channels reveal that the proposed autoencoder outperforms the Viterbi-based maximum-likelihood sequence estimation algorithm with perfect channel state information, when E b /N 0 takes low to medium values, i.e., less than 13 dB.
DOI:10.1109/GCWkshps50303.2020.9367429