SAS-SEINet: A SNR-Aware Adaptive Scalable SEI Neural Network Accelerator Using Algorithm–Hardware Co-Design for High-Accuracy and Power-Efficient UAV Surveillance
As a potential air control measure, RF-based surveillance is one of the most commonly used unmanned aerial vehicles (UAV) surveillance methods that exploits specific emitter identification (SEI) technology to identify captured RF signal from ground controllers to UAVs. Recently many SEI algorithms b...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-08-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/17/6532 |
| _version_ | 1797493084579168256 |
|---|---|
| author | Jiayan Gan Ang Hu Ziyi Kang Zhipeng Qu Zhanxiang Yang Rui Yang Yibing Wang Huaizong Shao Jun Zhou |
| author_facet | Jiayan Gan Ang Hu Ziyi Kang Zhipeng Qu Zhanxiang Yang Rui Yang Yibing Wang Huaizong Shao Jun Zhou |
| author_sort | Jiayan Gan |
| collection | DOAJ |
| description | As a potential air control measure, RF-based surveillance is one of the most commonly used unmanned aerial vehicles (UAV) surveillance methods that exploits specific emitter identification (SEI) technology to identify captured RF signal from ground controllers to UAVs. Recently many SEI algorithms based on deep convolution neural network (DCNN) have emerged. However, there is a lack of the implementation of specific hardware. This paper proposes a high-accuracy and power-efficient hardware accelerator using an algorithm–hardware co-design for UAV surveillance. For the algorithm, we propose a scalable SEI neural network with SNR-aware adaptive precision computation. With SNR awareness and precision reconfiguration, it can adaptively switch between DCNN and binary DCNN to cope with low SNR and high SNR tasks, respectively. In addition, a short-time Fourier transform (STFT) reusing DCNN method is proposed to pre-extract feature of UAV signal. For hardware, we designed a SNR sensing engine, denoising engine, and specialized DCNN engine with hybrid-precision convolution and memory access, aiming at SEI acceleration. Finally, we validate the effectiveness of our design on a FPGA, using a public UAV dataset. Compared with a state-of-the-art algorithm, our method can achieve the highest accuracy of 99.3% and an F1 score of 99.3%. Compared with other hardware designs, our accelerator can achieve the highest power efficiency of 40.12 Gops/W and 96.52 Gops/W with INT16 precision and binary precision. |
| first_indexed | 2024-03-10T01:14:56Z |
| format | Article |
| id | doaj.art-2fa6d1f3f39644cfb7fd2cc998c7cfe2 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T01:14:56Z |
| publishDate | 2022-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-2fa6d1f3f39644cfb7fd2cc998c7cfe22023-11-23T14:10:04ZengMDPI AGSensors1424-82202022-08-012217653210.3390/s22176532SAS-SEINet: A SNR-Aware Adaptive Scalable SEI Neural Network Accelerator Using Algorithm–Hardware Co-Design for High-Accuracy and Power-Efficient UAV SurveillanceJiayan Gan0Ang Hu1Ziyi Kang2Zhipeng Qu3Zhanxiang Yang4Rui Yang5Yibing Wang6Huaizong Shao7Jun Zhou8School of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu 611731, ChinaSchool of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu 611731, ChinaSchool of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu 611731, ChinaSchool of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu 611731, ChinaSchool of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu 611731, ChinaSchool of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu 611731, ChinaSchool of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu 611731, ChinaSchool of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu 611731, ChinaSchool of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu 611731, ChinaAs a potential air control measure, RF-based surveillance is one of the most commonly used unmanned aerial vehicles (UAV) surveillance methods that exploits specific emitter identification (SEI) technology to identify captured RF signal from ground controllers to UAVs. Recently many SEI algorithms based on deep convolution neural network (DCNN) have emerged. However, there is a lack of the implementation of specific hardware. This paper proposes a high-accuracy and power-efficient hardware accelerator using an algorithm–hardware co-design for UAV surveillance. For the algorithm, we propose a scalable SEI neural network with SNR-aware adaptive precision computation. With SNR awareness and precision reconfiguration, it can adaptively switch between DCNN and binary DCNN to cope with low SNR and high SNR tasks, respectively. In addition, a short-time Fourier transform (STFT) reusing DCNN method is proposed to pre-extract feature of UAV signal. For hardware, we designed a SNR sensing engine, denoising engine, and specialized DCNN engine with hybrid-precision convolution and memory access, aiming at SEI acceleration. Finally, we validate the effectiveness of our design on a FPGA, using a public UAV dataset. Compared with a state-of-the-art algorithm, our method can achieve the highest accuracy of 99.3% and an F1 score of 99.3%. Compared with other hardware designs, our accelerator can achieve the highest power efficiency of 40.12 Gops/W and 96.52 Gops/W with INT16 precision and binary precision.https://www.mdpi.com/1424-8220/22/17/6532UAVSEIDCNNSNRpower efficiency |
| spellingShingle | Jiayan Gan Ang Hu Ziyi Kang Zhipeng Qu Zhanxiang Yang Rui Yang Yibing Wang Huaizong Shao Jun Zhou SAS-SEINet: A SNR-Aware Adaptive Scalable SEI Neural Network Accelerator Using Algorithm–Hardware Co-Design for High-Accuracy and Power-Efficient UAV Surveillance Sensors UAV SEI DCNN SNR power efficiency |
| title | SAS-SEINet: A SNR-Aware Adaptive Scalable SEI Neural Network Accelerator Using Algorithm–Hardware Co-Design for High-Accuracy and Power-Efficient UAV Surveillance |
| title_full | SAS-SEINet: A SNR-Aware Adaptive Scalable SEI Neural Network Accelerator Using Algorithm–Hardware Co-Design for High-Accuracy and Power-Efficient UAV Surveillance |
| title_fullStr | SAS-SEINet: A SNR-Aware Adaptive Scalable SEI Neural Network Accelerator Using Algorithm–Hardware Co-Design for High-Accuracy and Power-Efficient UAV Surveillance |
| title_full_unstemmed | SAS-SEINet: A SNR-Aware Adaptive Scalable SEI Neural Network Accelerator Using Algorithm–Hardware Co-Design for High-Accuracy and Power-Efficient UAV Surveillance |
| title_short | SAS-SEINet: A SNR-Aware Adaptive Scalable SEI Neural Network Accelerator Using Algorithm–Hardware Co-Design for High-Accuracy and Power-Efficient UAV Surveillance |
| title_sort | sas seinet a snr aware adaptive scalable sei neural network accelerator using algorithm hardware co design for high accuracy and power efficient uav surveillance |
| topic | UAV SEI DCNN SNR power efficiency |
| url | https://www.mdpi.com/1424-8220/22/17/6532 |
| work_keys_str_mv | AT jiayangan sasseinetasnrawareadaptivescalableseineuralnetworkacceleratorusingalgorithmhardwarecodesignforhighaccuracyandpowerefficientuavsurveillance AT anghu sasseinetasnrawareadaptivescalableseineuralnetworkacceleratorusingalgorithmhardwarecodesignforhighaccuracyandpowerefficientuavsurveillance AT ziyikang sasseinetasnrawareadaptivescalableseineuralnetworkacceleratorusingalgorithmhardwarecodesignforhighaccuracyandpowerefficientuavsurveillance AT zhipengqu sasseinetasnrawareadaptivescalableseineuralnetworkacceleratorusingalgorithmhardwarecodesignforhighaccuracyandpowerefficientuavsurveillance AT zhanxiangyang sasseinetasnrawareadaptivescalableseineuralnetworkacceleratorusingalgorithmhardwarecodesignforhighaccuracyandpowerefficientuavsurveillance AT ruiyang sasseinetasnrawareadaptivescalableseineuralnetworkacceleratorusingalgorithmhardwarecodesignforhighaccuracyandpowerefficientuavsurveillance AT yibingwang sasseinetasnrawareadaptivescalableseineuralnetworkacceleratorusingalgorithmhardwarecodesignforhighaccuracyandpowerefficientuavsurveillance AT huaizongshao sasseinetasnrawareadaptivescalableseineuralnetworkacceleratorusingalgorithmhardwarecodesignforhighaccuracyandpowerefficientuavsurveillance AT junzhou sasseinetasnrawareadaptivescalableseineuralnetworkacceleratorusingalgorithmhardwarecodesignforhighaccuracyandpowerefficientuavsurveillance |