FE-SKViT: A Feature-Enhanced ViT Model with Skip Attention for Automatic Modulation Recognition
Automatic modulation recognition (AMR) is widely employed in communication systems. However, under conditions of low signal-to-noise ratio (SNR), recent studies reveal limitations in achieving high AMR accuracy. In this work, we introduce a novel network architecture that leverages a transformer-ins...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-11-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/16/22/4204 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850147199556517888 |
|---|---|
| author | Guangyao Zheng Bo Zang Penghui Yang Wenbo Zhang Bin Li |
| author_facet | Guangyao Zheng Bo Zang Penghui Yang Wenbo Zhang Bin Li |
| author_sort | Guangyao Zheng |
| collection | DOAJ |
| description | Automatic modulation recognition (AMR) is widely employed in communication systems. However, under conditions of low signal-to-noise ratio (SNR), recent studies reveal limitations in achieving high AMR accuracy. In this work, we introduce a novel network architecture that leverages a transformer-inspired approach tailored for AMR, called Feature-Enhanced Transformer with skip-attention (FE-SKViT). This innovative design adeptly harnesses the advantages of translation variant convolution and the Transformer framework, handling intra-signal variance and small cross-signal variance to achieve enhanced recognition accuracy. Experimental results on RadioML2016.10a, RadioML2016.10b, and RML22 datasets demonstrate that the Feature-Enhanced Transformer with skip-attention (FE-SKViT) excels over other methods, particularly under low SNR conditions ranging from −4 to 6 dB. |
| format | Article |
| id | doaj-art-cca8a80adfca4569b2224775781fee8b |
| institution | OA Journals |
| issn | 2072-4292 |
| language | English |
| publishDate | 2024-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj-art-cca8a80adfca4569b2224775781fee8b2025-08-20T02:27:38ZengMDPI AGRemote Sensing2072-42922024-11-011622420410.3390/rs16224204FE-SKViT: A Feature-Enhanced ViT Model with Skip Attention for Automatic Modulation RecognitionGuangyao Zheng0Bo Zang1Penghui Yang2Wenbo Zhang3Bin Li4School of Electronic Engineering, Xidian University, Xi’an 710071, ChinaSchool of Electronic Engineering, Xidian University, Xi’an 710071, ChinaChina Xi’an Satellite Control Center, Xi’an 710043, ChinaSchool of Electronic Engineering, Xidian University, Xi’an 710071, ChinaChina Xi’an Satellite Control Center, Xi’an 710043, ChinaAutomatic modulation recognition (AMR) is widely employed in communication systems. However, under conditions of low signal-to-noise ratio (SNR), recent studies reveal limitations in achieving high AMR accuracy. In this work, we introduce a novel network architecture that leverages a transformer-inspired approach tailored for AMR, called Feature-Enhanced Transformer with skip-attention (FE-SKViT). This innovative design adeptly harnesses the advantages of translation variant convolution and the Transformer framework, handling intra-signal variance and small cross-signal variance to achieve enhanced recognition accuracy. Experimental results on RadioML2016.10a, RadioML2016.10b, and RML22 datasets demonstrate that the Feature-Enhanced Transformer with skip-attention (FE-SKViT) excels over other methods, particularly under low SNR conditions ranging from −4 to 6 dB.https://www.mdpi.com/2072-4292/16/22/4204automatic modulation recognitionvision transformerstranslation variant convolution |
| spellingShingle | Guangyao Zheng Bo Zang Penghui Yang Wenbo Zhang Bin Li FE-SKViT: A Feature-Enhanced ViT Model with Skip Attention for Automatic Modulation Recognition Remote Sensing automatic modulation recognition vision transformers translation variant convolution |
| title | FE-SKViT: A Feature-Enhanced ViT Model with Skip Attention for Automatic Modulation Recognition |
| title_full | FE-SKViT: A Feature-Enhanced ViT Model with Skip Attention for Automatic Modulation Recognition |
| title_fullStr | FE-SKViT: A Feature-Enhanced ViT Model with Skip Attention for Automatic Modulation Recognition |
| title_full_unstemmed | FE-SKViT: A Feature-Enhanced ViT Model with Skip Attention for Automatic Modulation Recognition |
| title_short | FE-SKViT: A Feature-Enhanced ViT Model with Skip Attention for Automatic Modulation Recognition |
| title_sort | fe skvit a feature enhanced vit model with skip attention for automatic modulation recognition |
| topic | automatic modulation recognition vision transformers translation variant convolution |
| url | https://www.mdpi.com/2072-4292/16/22/4204 |
| work_keys_str_mv | AT guangyaozheng feskvitafeatureenhancedvitmodelwithskipattentionforautomaticmodulationrecognition AT bozang feskvitafeatureenhancedvitmodelwithskipattentionforautomaticmodulationrecognition AT penghuiyang feskvitafeatureenhancedvitmodelwithskipattentionforautomaticmodulationrecognition AT wenbozhang feskvitafeatureenhancedvitmodelwithskipattentionforautomaticmodulationrecognition AT binli feskvitafeatureenhancedvitmodelwithskipattentionforautomaticmodulationrecognition |