Automatic Signal Modulation Recognition based on Deep Convolutional Neural Network

DOI

10.2991/iccia-19.2019.86How to use a DOI?

Keywords

deep learning (DL); neural network; modulation; wireless communication.

Abstract

Deep learning (DL) shows great vitality in all areas, but it rarely involves wireless communication. This paper proposes an automatic signal modulation recognition method based on deep convolutional neural network to solve common problems in wireless communication. The algorithm automatically extracts various feature details of the image through the deep convolutional neural network of deep learning, instead of the huge engineering of manual design features to achieve accurate recognition of signal and noise under various signal-to-noise ratio conditions. The method uses the image processing GPU to build VGGNet to automatically recognize 10 kinds of modulated signals in MPSK and MQAM under the deep learning architecture TensorFlow. The simulation results show that the minimum recognition accuracy of various signals is 96.7% when the signal-to-noise ratio is 5dB. Compared with other methods, the proposed method is better.

Copyright

© 2019, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Yulin Sun
AU  - Jun Li
AU  - Fei Lin
AU  - Guangliang Pan
PY  - 2019/07
DA  - 2019/07
TI  - Automatic Signal Modulation Recognition based on Deep Convolutional Neural Network
BT  - Proceedings of the 3rd International Conference on Computer Engineering, Information Science & Application Technology (ICCIA 2019)
PB  - Atlantis Press
SP  - 550
EP  - 554
SN  - 2352-538X
UR  - https://doi.org/10.2991/iccia-19.2019.86
DO  - 10.2991/iccia-19.2019.86
ID  - Sun2019/07
ER  -