An efficient CNN-RNN recognition network for complex interference signal

Zhuangxin Fang; Zhiyong Luo; Xiti Wang

doi:10.1117/12.2667930

28 June 2023 An efficient CNN-RNN recognition network for complex interference signal

Zhuangxin Fang, Zhiyong Luo, Xiti Wang

Proceedings Volume 12720, 2022 Workshop on Electronics Communication Engineering; 127200E (2023) https://doi.org/10.1117/12.2667930
Event: 2022 Workshop on Electronics Communication Engineering (WECE 2022), 2022, Xi'an, China

Abstract

Facing an increasingly complex electromagnetic environment, modern communication systems must adopt certain antiinterference technology when deploying system equipment and network to ensure the normal operation of wireless communication. Currently, interference recognition is the foundation and key link of anti-interference technology. Among them, the recognition accuracy and the dependence of the algorithm model on training data are challenges that need to be solved urgently. In this paper, a CNN-RNN joint network architecture combining residual network and LSTM network is proposed to recognize the interfering signals. The joint network architecture adopts the parallel combination of residual and LSTM network, where the time-frequency image data of signals is input to the residual network branches while the real part, imaginary part, and spectral amplitude data of signals are input to the LSTM network branches. After simulation verification, the interference recognition result of the joint network is significantly improved compared with the single network. Firstly, compared with the single LSTM network, even though the single LSTM network has reached a very high recognition accuracy, the recognition accuracy of the joint network is still about 1%∼2% higher. What’s more, compared with the single network, the interference noise ratio (INR) generalization ability of the joint network is obviously improved. After training the network with different INR distributions, the recognition accuracy can be maintained. Therefore, it’s not sensitive to the INR distribution of the training data, which can adapt to different distribution conditions of training data and reduces the dependence of the algorithm on training data.

Citation Download Citation

Zhuangxin Fang, Zhiyong Luo, and Xiti Wang "An efficient CNN-RNN recognition network for complex interference signal", Proc. SPIE 12720, 2022 Workshop on Electronics Communication Engineering, 127200E (28 June 2023); https://doi.org/10.1117/12.2667930

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
12 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Education and training

Feature extraction

Network architectures

Time-frequency analysis

Detection and tracking algorithms

Signal processing

Image processing

Show All Keywords

Keywords/Phrases

Search In:

Publication Years