Visualizing Deep Learning-based Radio Modulation Classifier

TL;DR

This paper visualizes deep learning-based radio modulation classifiers, revealing that CNN and LSTM models extract similar features related to modulation points, with LSTM features aligning with expert knowledge, but short samples can cause misclassification.

Contribution

It introduces a class activation vector for visualizing features in deep learning radio classifiers, comparing CNN and LSTM models and analyzing their interpretability.

Findings

CNN and LSTM extract similar modulation-related features

LSTM features align with human expert knowledge

Short radio samples can lead to misclassification

Abstract

Deep learning has recently been successfully applied in automatic modulation classification by extracting and classifying radio features in an end-to-end way. However, deep learning-based radio modulation classifiers are lack of interpretability, and there is little explanation or visibility into what kinds of radio features are extracted and chosen for classification. In this paper, we visualize different deep learning-based radio modulation classifiers by introducing a class activation vector. Specifically, both convolutional neural networks (CNN) based classifier and long short-term memory (LSTM) based classifier are separately studied, and their extracted radio features are visualized. Extensive numerical results show both the CNN-based classifier and LSTM-based classifier extract similar radio features relating to modulation reference points. In particular, for the LSTM-based classifier, its obtained radio features are similar to the knowledge of human experts. Our numerical results indicate the radio features extracted by deep learning-based classifiers greatly depend on the contents carried by radio signals, and a short radio sample may lead to misclassification.