Space Target Classification Improvement by Generating Micro-Doppler Signatures Considering Incident Angle
Classifying space targets from debris is critical for radar resource management as well as rapid response during the mid-course phase of space target flight. Due to advances in deep learning techniques, various approaches have been studied to classify space targets by using micro-Doppler signatures....
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-02-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/4/1653 |
| _version_ | 1797476678022201344 |
|---|---|
| author | Jae-In Lee Nammon Kim Sawon Min Jeongwoo Kim Dae-Kyo Jeong Dong-Wook Seo |
| author_facet | Jae-In Lee Nammon Kim Sawon Min Jeongwoo Kim Dae-Kyo Jeong Dong-Wook Seo |
| author_sort | Jae-In Lee |
| collection | DOAJ |
| description | Classifying space targets from debris is critical for radar resource management as well as rapid response during the mid-course phase of space target flight. Due to advances in deep learning techniques, various approaches have been studied to classify space targets by using micro-Doppler signatures. Previous studies have only used micro-Doppler signatures such as spectrogram and cadence velocity diagram (CVD), but in this paper, we propose a method to generate micro-Doppler signatures taking into account the relative incident angle that a radar can obtain during the target tracking process. The AlexNet and ResNet-18 networks, which are representative convolutional neural network architectures, are transfer-learned using two types of datasets constructed using the proposed and conventional signatures to classify six classes of space targets and a debris–cone, rounded cone, cone with empennages, cylinder, curved plate, and square plate. Among the proposed signatures, the spectrogram had lower classification accuracy than the conventional spectrogram, but the classification accuracy increased from 88.97% to 92.11% for CVD. Furthermore, when recalculated not with six classes but simply with only two classes of precessing space targets and tumbling debris, the proposed spectrogram and CVD show the classification accuracy of over 99.82% for both AlexNet and ResNet-18. Specially, for two classes, CVD provided results with higher accuracy than the spectrogram. |
| first_indexed | 2024-03-09T21:05:02Z |
| format | Article |
| id | doaj.art-47a28e09488e4003b6b3e314163a26a4 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T21:05:02Z |
| publishDate | 2022-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-47a28e09488e4003b6b3e314163a26a42023-11-23T22:03:02ZengMDPI AGSensors1424-82202022-02-01224165310.3390/s22041653Space Target Classification Improvement by Generating Micro-Doppler Signatures Considering Incident AngleJae-In Lee0Nammon Kim1Sawon Min2Jeongwoo Kim3Dae-Kyo Jeong4Dong-Wook Seo5Interdisciplinary Major of Maritime AI Convergence, Korea Maritime & Ocean University, Busan 49112, KoreaDepartment of Land Radar, Hanwha Systems, Yongin 17121, KoreaDepartment of Land Radar, Hanwha Systems, Yongin 17121, KoreaDepartment of Land Radar, Hanwha Systems, Yongin 17121, KoreaAgency for Defense Development, Daejeon 34075, KoreaInterdisciplinary Major of Maritime AI Convergence, Korea Maritime & Ocean University, Busan 49112, KoreaClassifying space targets from debris is critical for radar resource management as well as rapid response during the mid-course phase of space target flight. Due to advances in deep learning techniques, various approaches have been studied to classify space targets by using micro-Doppler signatures. Previous studies have only used micro-Doppler signatures such as spectrogram and cadence velocity diagram (CVD), but in this paper, we propose a method to generate micro-Doppler signatures taking into account the relative incident angle that a radar can obtain during the target tracking process. The AlexNet and ResNet-18 networks, which are representative convolutional neural network architectures, are transfer-learned using two types of datasets constructed using the proposed and conventional signatures to classify six classes of space targets and a debris–cone, rounded cone, cone with empennages, cylinder, curved plate, and square plate. Among the proposed signatures, the spectrogram had lower classification accuracy than the conventional spectrogram, but the classification accuracy increased from 88.97% to 92.11% for CVD. Furthermore, when recalculated not with six classes but simply with only two classes of precessing space targets and tumbling debris, the proposed spectrogram and CVD show the classification accuracy of over 99.82% for both AlexNet and ResNet-18. Specially, for two classes, CVD provided results with higher accuracy than the spectrogram.https://www.mdpi.com/1424-8220/22/4/1653classificationconvolution neural networkdebrisdeep learningincident anglemicro-Doppler |
| spellingShingle | Jae-In Lee Nammon Kim Sawon Min Jeongwoo Kim Dae-Kyo Jeong Dong-Wook Seo Space Target Classification Improvement by Generating Micro-Doppler Signatures Considering Incident Angle Sensors classification convolution neural network debris deep learning incident angle micro-Doppler |
| title | Space Target Classification Improvement by Generating Micro-Doppler Signatures Considering Incident Angle |
| title_full | Space Target Classification Improvement by Generating Micro-Doppler Signatures Considering Incident Angle |
| title_fullStr | Space Target Classification Improvement by Generating Micro-Doppler Signatures Considering Incident Angle |
| title_full_unstemmed | Space Target Classification Improvement by Generating Micro-Doppler Signatures Considering Incident Angle |
| title_short | Space Target Classification Improvement by Generating Micro-Doppler Signatures Considering Incident Angle |
| title_sort | space target classification improvement by generating micro doppler signatures considering incident angle |
| topic | classification convolution neural network debris deep learning incident angle micro-Doppler |
| url | https://www.mdpi.com/1424-8220/22/4/1653 |
| work_keys_str_mv | AT jaeinlee spacetargetclassificationimprovementbygeneratingmicrodopplersignaturesconsideringincidentangle AT nammonkim spacetargetclassificationimprovementbygeneratingmicrodopplersignaturesconsideringincidentangle AT sawonmin spacetargetclassificationimprovementbygeneratingmicrodopplersignaturesconsideringincidentangle AT jeongwookim spacetargetclassificationimprovementbygeneratingmicrodopplersignaturesconsideringincidentangle AT daekyojeong spacetargetclassificationimprovementbygeneratingmicrodopplersignaturesconsideringincidentangle AT dongwookseo spacetargetclassificationimprovementbygeneratingmicrodopplersignaturesconsideringincidentangle |