Summary

International Conference on Emerging Technologies for Communications

2020

Session Number:I1

Session:

Number:I1-5

A Preamble-less Symbol Timing Synchronization for OFDM Systems Using Convolutional Neural Networks

Yuta GOTO,  Shun KOJIMA,  Kakeru MATSUBARA,  Naoya TANUMA,  Koki ITO,  Kazuki MARUTA,  Chang-Jun AHN,  

pp.-

Publication Date:2020/12/2

Online ISSN:2188-5079

DOI:10.34385/proc.63.I1-5

PDF download

PayPerView

Summary:
This article proposes a preamble-less symbol timing synchronization method for OFDM systems using convolutional neural networks (CNN). CNN learns the characteristics of spectra and helps in synchronization. Simulation experiment shows that the proposed method is superior to the conventional method in the high Eb/N0 region.