Time-Frequency Analysis and Target Recognition of HRRP Based on CN-LSGAN, STFT, and CNN
Aiming at the problem of radar target recognition of High-Resolution Range Profile (HRRP) under low signal-to-noise ratio conditions, a recognition method based on the Constrained Naive Least-Squares Generative Adversarial Network (CN-LSGAN), Short-time Fourier Transform (STFT), and Convolutional Ne...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2021-01-01
|
| Series: | Complexity |
| Online Access: | http://dx.doi.org/10.1155/2021/6664530 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832566439341457408 |
|---|---|
| author | Jianghua Nie Yongsheng Xiao Lizhen Huang Feng Lv |
| author_facet | Jianghua Nie Yongsheng Xiao Lizhen Huang Feng Lv |
| author_sort | Jianghua Nie |
| collection | DOAJ |
| description | Aiming at the problem of radar target recognition of High-Resolution Range Profile (HRRP) under low signal-to-noise ratio conditions, a recognition method based on the Constrained Naive Least-Squares Generative Adversarial Network (CN-LSGAN), Short-time Fourier Transform (STFT), and Convolutional Neural Network (CNN) is proposed. Combining the Least-Squares Generative Adversarial Network (LSGAN) with the Wasserstein Generative Adversarial Network with Gradient Penalty (WGAN-GP), the CN-LSGAN is presented and applied to the HRRP denoise. The frequency domain and phase features of HRRP are gained by STFT in order to facilitate feature learning and also match the input data format of the CNN. These experimental results show that the CN-LSGAN has better data augmentation performance and can effectively avoid the model collapse compared to the generative adversarial network (GAN) and LSGAN. Also, the method has better recognition performance than the one-dimensional CNN method and the Long Short-Term Memory (LSTM) network method. |
| format | Article |
| id | doaj-art-cae94037a9bc40ac98c77ed8801c5dd7 |
| institution | Kabale University |
| issn | 1076-2787 1099-0526 |
| language | English |
| publishDate | 2021-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | Complexity |
| spelling | doaj-art-cae94037a9bc40ac98c77ed8801c5dd72025-02-03T01:04:09ZengWileyComplexity1076-27871099-05262021-01-01202110.1155/2021/66645306664530Time-Frequency Analysis and Target Recognition of HRRP Based on CN-LSGAN, STFT, and CNNJianghua Nie0Yongsheng Xiao1Lizhen Huang2Feng Lv3Key Laboratory of Jiangxi Province for Image Processing and Pattern Recognition, School of Information Engineering, Nanchang Hangkong University, Nanchang 330063, ChinaKey Laboratory of Jiangxi Province for Image Processing and Pattern Recognition, School of Information Engineering, Nanchang Hangkong University, Nanchang 330063, ChinaKey Laboratory of Jiangxi Province for Image Processing and Pattern Recognition, School of Information Engineering, Nanchang Hangkong University, Nanchang 330063, ChinaHenan Aerospace Hydraulic Pneumatic Technology Company Limited, China Aerospace Science and Industry Corporation Limited, Zhengzhou 451191, ChinaAiming at the problem of radar target recognition of High-Resolution Range Profile (HRRP) under low signal-to-noise ratio conditions, a recognition method based on the Constrained Naive Least-Squares Generative Adversarial Network (CN-LSGAN), Short-time Fourier Transform (STFT), and Convolutional Neural Network (CNN) is proposed. Combining the Least-Squares Generative Adversarial Network (LSGAN) with the Wasserstein Generative Adversarial Network with Gradient Penalty (WGAN-GP), the CN-LSGAN is presented and applied to the HRRP denoise. The frequency domain and phase features of HRRP are gained by STFT in order to facilitate feature learning and also match the input data format of the CNN. These experimental results show that the CN-LSGAN has better data augmentation performance and can effectively avoid the model collapse compared to the generative adversarial network (GAN) and LSGAN. Also, the method has better recognition performance than the one-dimensional CNN method and the Long Short-Term Memory (LSTM) network method.http://dx.doi.org/10.1155/2021/6664530 |
| spellingShingle | Jianghua Nie Yongsheng Xiao Lizhen Huang Feng Lv Time-Frequency Analysis and Target Recognition of HRRP Based on CN-LSGAN, STFT, and CNN Complexity |
| title | Time-Frequency Analysis and Target Recognition of HRRP Based on CN-LSGAN, STFT, and CNN |
| title_full | Time-Frequency Analysis and Target Recognition of HRRP Based on CN-LSGAN, STFT, and CNN |
| title_fullStr | Time-Frequency Analysis and Target Recognition of HRRP Based on CN-LSGAN, STFT, and CNN |
| title_full_unstemmed | Time-Frequency Analysis and Target Recognition of HRRP Based on CN-LSGAN, STFT, and CNN |
| title_short | Time-Frequency Analysis and Target Recognition of HRRP Based on CN-LSGAN, STFT, and CNN |
| title_sort | time frequency analysis and target recognition of hrrp based on cn lsgan stft and cnn |
| url | http://dx.doi.org/10.1155/2021/6664530 |
| work_keys_str_mv | AT jianghuanie timefrequencyanalysisandtargetrecognitionofhrrpbasedoncnlsganstftandcnn AT yongshengxiao timefrequencyanalysisandtargetrecognitionofhrrpbasedoncnlsganstftandcnn AT lizhenhuang timefrequencyanalysisandtargetrecognitionofhrrpbasedoncnlsganstftandcnn AT fenglv timefrequencyanalysisandtargetrecognitionofhrrpbasedoncnlsganstftandcnn |