Robust approach for AMC in frequency selective fading scenarios using unsupervised sparse-autoencoder-based deep neural network

Application of deep learning in the area of automatic modulation classification (AMC) is still evolving. An unsupervised sparse-autoencoder-based deep neural network (SAE-DNN) is proposed to deal with the problem of AMC for much neglected frequency selective fading scenarios with Doppler shift. The...

Full description

Saved in:
Bibliographic Details
Published in:IET communications Vol. 13; no. 4; pp. 423 - 432
Main Authors: Shah, Maqsood Hussain, Dang, Xiaoyu
Format: Journal Article
Language:English
Published: The Institution of Engineering and Technology 05.03.2019
Subjects:
AMC
cumulant based features
deep learning
Doppler shift
feature extraction
higher order statistics
input training data
learning (artificial intelligence)
low complexity solution
low complexity spectral
modulation
neglected frequency selective fading scenarios
neural nets
refined features
Research Article
robust approach
robust features
signal classification
training SAE‐DNN
unique training method
unsupervised sparse‐autoencoder‐based deep neural network
input training data
modulation
unique training method
neglected frequency selective fading scenarios
higher order statistics
training SAE-DNN
signal classification
Doppler shift
deep learning
low complexity spectral
feature extraction
AMC
robust approach
refined features
cumulant based features
robust features
unsupervised sparse-autoencoder-based deep neural network
learning (artificial intelligence)
low complexity solution
neural nets
ISSN:1751-8628, 1751-8636
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Application of deep learning in the area of automatic modulation classification (AMC) is still evolving. An unsupervised sparse-autoencoder-based deep neural network (SAE-DNN) is proposed to deal with the problem of AMC for much neglected frequency selective fading scenarios with Doppler shift. The authors propose a set of low complexity spectral and cumulant based features for training SAE-DNN. The network is designed using forced dimensionality reduction and sparsity constraint to achieve a low complexity solution with improved ability to learn more refined and robust features from the input training data. A unique training method is presented in this study which incorporates a range of SNR values for the entire span of the training dataset, as compared to the conventional approach which only uses a single SNR value for all the training examples. A comprehensive performance analysis shows that the proposed method outperforms many conventional counterparts in the literature. Generalisation test verifies that network is feasible for all channel conditions. A robust classification behaviour is observed against phase-frequency impairments and Doppler shift for frequency selective fading scenarios.
ISSN:1751-8628
1751-8636
DOI:10.1049/iet-com.2018.5688