Robust Bayesian attention belief network for radar work mode recognition

Understanding and analyzing radar work modes play a key role in electronic support measure system. Many classifiers, for example those based on convolutional neural network (CNN) and recurrent neural network (RNN), are available for recognizing radar work modes as well as emitter types from their waveform parameters. However, the performance of these methods may suffer significantly when confronting different types of signal degradation, e.g., measurement error, lost pulse and spurious pulse. To tackle this issue, we in this paper develop a Bayesian attention belief network (BABNet) based on Bayesian neural networks in which the probability distribution over weights can help to enhance the model robustness for corrupted data. In particular, we adopt pre-trained CNN as the Bayesian inference prior. This not only accelerates the convergence speed, but also avoids the training process getting stuck in bad local minima. Meanwhile, instead of using RNNs which are difficult to be implemented in parallel, the combination of padding operation and attention module in the proposed BABNet enables CNN, as the backbone, to process sequential data with variable length. Extensive experiments are conducted to demonstrate the recognition capability and robustness of the BABNet in different environments.

Attention mechanism, Bayesian neural network, Pulse descriptor word, Radar work mode, Recognition, Robustness

10.1016/j.dsp.2022.103874

0960-3174

Du, Mingyang

d42b1519-40d9-476a-b2a3-2b9f6c63d46a

Zhong, Ping

a51c7669-d590-4a7b-adc8-0f7d2b9e1932

Cai, Xiaohao

de483445-45e9-4b21-a4e8-b0427fc72cee

Bi, Daping

7d0942b5-14a9-4e09-b709-416fa34f31a2

Jing, Aiqi

c303e8cf-3555-4b90-aa8a-fab8803280f5

1 March 2023

Du, Mingyang

d42b1519-40d9-476a-b2a3-2b9f6c63d46a

Zhong, Ping

a51c7669-d590-4a7b-adc8-0f7d2b9e1932

Cai, Xiaohao

de483445-45e9-4b21-a4e8-b0427fc72cee

Bi, Daping

7d0942b5-14a9-4e09-b709-416fa34f31a2

Jing, Aiqi

c303e8cf-3555-4b90-aa8a-fab8803280f5

Du, Mingyang, Zhong, Ping, Cai, Xiaohao, Bi, Daping and Jing, Aiqi (2023) Robust Bayesian attention belief network for radar work mode recognition. Statistics and Computing, 133, [103874]. (doi:10.1016/j.dsp.2022.103874).

Record type: Article

Abstract

Text

Robust Bayesian Attention Belief Network for Radar Work Mode Recognition - Accepted Manuscript

Restricted to Repository staff only until 6 December 2024.

Available under License Creative Commons Attribution Non-commercial No Derivatives.

Request a copy

More information

e-pub ahead of print date: 7 December 2022

Published date: 1 March 2023

Additional Information: Funding Information: Ping ZHONG (Senior Member, IEEE) received the M.S. degree in applied mathematics and the Ph.D. degree in information and communication engineering from the National University of Defense Technology (NUDT), Changsha, China, in 2003 and 2008, respectively. Dr. Zhong was a recipient of the National Excellent Doctoral Dissertation Award of China in 2011 and the New Century Excellent Talents in the University of China in 2013. From March 2015 to February 2016, he was a visiting Scholar with the Department of Applied Mathematics and Theory Physics, University of Cambridge, Cambridge, U.K. He is currently a Professor with the National Key Laboratory of Science and Technology on ATR, NUDT. He has authored more than 40 peer-reviewed articles in international journals, such as the IEEE transactions and letters. His research interests include computer vision, machine learning and pattern recognition. Funding Information: This document is the results of the research project funded by the National Natural Science Foundation of China (Grant no. 61971428 ). Publisher Copyright: © 2022 Elsevier Inc.

Keywords: Attention mechanism, Bayesian neural network, Pulse descriptor word, Radar work mode, Recognition, Robustness

Learn more about Vision, Learning and Control research