Adaptive intra‐pulse interference waveform generation technique based on Convolutional Neural Network—Deep Neural Network with global priority information of radar signal

Abstract To adapt to the complex and changeable electromagnetic environment of radar detection, improve the jamming effect on frequency‐agile radar and low probability of intercept radar, and solve the problem of poor jamming effect caused by intra‐pulse jamming lagging behind target radar signal, a...

Full description

Bibliographic Details
Main Authors: Yilin Jiang, Xi Shang, Lisong Guan, Jinxin Li
Format: Article
Language:English
Published: Wiley 2023-02-01
Series:IET Radar, Sonar & Navigation
Online Access:https://doi.org/10.1049/rsn2.12334
_version_ 1797905486083784704
author Yilin Jiang
Xi Shang
Lisong Guan
Jinxin Li
author_facet Yilin Jiang
Xi Shang
Lisong Guan
Jinxin Li
author_sort Yilin Jiang
collection DOAJ
description Abstract To adapt to the complex and changeable electromagnetic environment of radar detection, improve the jamming effect on frequency‐agile radar and low probability of intercept radar, and solve the problem of poor jamming effect caused by intra‐pulse jamming lagging behind target radar signal, a Convolutional Neural Network—Deep Neural Network jamming waveform generation method, based on prior global information of radar signal, is presented in this study. This proposed method uses two networks to generate the overall interference waveforms. One Convolutional Neural Network—Deep Neural Network derives prior global radar information from local prior radar information, which is used by the other Convolutional Neural Network—Deep Neural Network to generate interference. In the whole scheme, an algorithm to design the intra‐pulse interference waveform based on the prior global information of the radar signal is proposed. The algorithm can generate the interference waveform samples from the number and area of multiple false targets and the relative peak between multiple false targets and the real target area by the relevant parameters. Then, an evaluation system (Matched filter, Constant False Alarm Rate Detector, and Distance resolution etc.) is established to evaluate the global effect of the designed interference waveforms. Finally, the time‐domain Root Mean Squared Errors is used as the loss function to optimise the training of the network and ultimately achieve the requirement of generating intra‐pulse adaptive interference waveforms based on radar signal fragments. The principle of generating interference waveforms is based on various aspects of the evaluation of the interference effect, which enhances the time domain correlation between multiple false target areas, thereby boosting the interference effect on the radar. The experimental results show that the proposed method based on the Convolutional Neural Network—Deep Neural Network can effectively solve the problem of intra‐pulse interference lag and small effective interference area after the pulse pressure of the interference waveform matching filter. This study offers certain reference significance for the adaptive interference waveform design based on radar prior global information.
first_indexed 2024-04-10T10:06:05Z
format Article
id doaj.art-cdffc5ac72c546cfbf6d4499586da46a
institution Directory Open Access Journal
issn 1751-8784
1751-8792
language English
last_indexed 2024-04-10T10:06:05Z
publishDate 2023-02-01
publisher Wiley
record_format Article
series IET Radar, Sonar & Navigation
spelling doaj.art-cdffc5ac72c546cfbf6d4499586da46a2023-02-15T18:03:35ZengWileyIET Radar, Sonar & Navigation1751-87841751-87922023-02-0117221222610.1049/rsn2.12334Adaptive intra‐pulse interference waveform generation technique based on Convolutional Neural Network—Deep Neural Network with global priority information of radar signalYilin Jiang0Xi Shang1Lisong Guan2Jinxin Li3College of Information and Communication Engineering Harbin Engineering University Harbin ChinaCollege of Information and Communication Engineering Harbin Engineering University Harbin ChinaCollege of Information and Communication Engineering Harbin Engineering University Harbin ChinaCollege of Information and Communication Engineering Harbin Engineering University Harbin ChinaAbstract To adapt to the complex and changeable electromagnetic environment of radar detection, improve the jamming effect on frequency‐agile radar and low probability of intercept radar, and solve the problem of poor jamming effect caused by intra‐pulse jamming lagging behind target radar signal, a Convolutional Neural Network—Deep Neural Network jamming waveform generation method, based on prior global information of radar signal, is presented in this study. This proposed method uses two networks to generate the overall interference waveforms. One Convolutional Neural Network—Deep Neural Network derives prior global radar information from local prior radar information, which is used by the other Convolutional Neural Network—Deep Neural Network to generate interference. In the whole scheme, an algorithm to design the intra‐pulse interference waveform based on the prior global information of the radar signal is proposed. The algorithm can generate the interference waveform samples from the number and area of multiple false targets and the relative peak between multiple false targets and the real target area by the relevant parameters. Then, an evaluation system (Matched filter, Constant False Alarm Rate Detector, and Distance resolution etc.) is established to evaluate the global effect of the designed interference waveforms. Finally, the time‐domain Root Mean Squared Errors is used as the loss function to optimise the training of the network and ultimately achieve the requirement of generating intra‐pulse adaptive interference waveforms based on radar signal fragments. The principle of generating interference waveforms is based on various aspects of the evaluation of the interference effect, which enhances the time domain correlation between multiple false target areas, thereby boosting the interference effect on the radar. The experimental results show that the proposed method based on the Convolutional Neural Network—Deep Neural Network can effectively solve the problem of intra‐pulse interference lag and small effective interference area after the pulse pressure of the interference waveform matching filter. This study offers certain reference significance for the adaptive interference waveform design based on radar prior global information.https://doi.org/10.1049/rsn2.12334
spellingShingle Yilin Jiang
Xi Shang
Lisong Guan
Jinxin Li
Adaptive intra‐pulse interference waveform generation technique based on Convolutional Neural Network—Deep Neural Network with global priority information of radar signal
IET Radar, Sonar & Navigation
title Adaptive intra‐pulse interference waveform generation technique based on Convolutional Neural Network—Deep Neural Network with global priority information of radar signal
title_full Adaptive intra‐pulse interference waveform generation technique based on Convolutional Neural Network—Deep Neural Network with global priority information of radar signal
title_fullStr Adaptive intra‐pulse interference waveform generation technique based on Convolutional Neural Network—Deep Neural Network with global priority information of radar signal
title_full_unstemmed Adaptive intra‐pulse interference waveform generation technique based on Convolutional Neural Network—Deep Neural Network with global priority information of radar signal
title_short Adaptive intra‐pulse interference waveform generation technique based on Convolutional Neural Network—Deep Neural Network with global priority information of radar signal
title_sort adaptive intra pulse interference waveform generation technique based on convolutional neural network deep neural network with global priority information of radar signal
url https://doi.org/10.1049/rsn2.12334
work_keys_str_mv AT yilinjiang adaptiveintrapulseinterferencewaveformgenerationtechniquebasedonconvolutionalneuralnetworkdeepneuralnetworkwithglobalpriorityinformationofradarsignal
AT xishang adaptiveintrapulseinterferencewaveformgenerationtechniquebasedonconvolutionalneuralnetworkdeepneuralnetworkwithglobalpriorityinformationofradarsignal
AT lisongguan adaptiveintrapulseinterferencewaveformgenerationtechniquebasedonconvolutionalneuralnetworkdeepneuralnetworkwithglobalpriorityinformationofradarsignal
AT jinxinli adaptiveintrapulseinterferencewaveformgenerationtechniquebasedonconvolutionalneuralnetworkdeepneuralnetworkwithglobalpriorityinformationofradarsignal