A convolutional and transformer based deep neural network for automatic modulation classification

• Published in: China communications Vol. 20; no. 5; pp. 135 - 147
• Main Authors: Ying, Shanchuan, Huang, Sai, Chang, Shuo, Yang, Zheng, Feng, Zhiyong, Guo, Ningyan
• Format: Journal Article
• Language: English
• Published: China Institute of Communications 01.05.2023; Key Laboratory of Universal Wireless Communications,Ministry of Education,Beijing University of Posts and Telecommunications,Beijing 100876,China
• Subjects: automatic modulation classification; Baseband; Convolution; convolutional neural network; Convolutional neural networks; deep neural network; Feature extraction; Kernel; Modulation; transformer; Transformers; convolutional neural network; automatic modulation classification; deep neural network; trans-former
• ISSN: 1673-5447
• DOI: 10.23919/JCC.ja.2022-0580

Automatic modulation classification (AMC) aims at identifying the modulation of the received signals, which is a significant approach to identifying the target in military and civil applications. In this paper, a novel data-driven framework named con-volutional and transformer-based deep neural network (CTDNN) is proposed to improve the classification performance. CTDNN can be divided into four modules, i.e., convolutional neural network (CNN) backbone, transition module, transformer module, and final classifier. In the CNN backbone, a wide and deep convolution structure is designed, which consists of 1 × 15 convolution kernels and intensive cross-layer connections instead of traditional 1×3 kernels and sequential connections. In the transition module, a 1 × 1 convolution layer is utilized to compress the channels of the previous multi-scale CNN features. In the transformer module, three self-attention layers are designed for extracting global features and generating the classification vector. In the classifier, the final decision is made based on the maximum a posterior probability. Extensive simulations are conducted, and the result shows that our proposed CTDNN can achieve superior classification performance than traditional deep models.