IEMEK Journal of Embedded Systems and Applications (대한임베디드공학회논문지)

Institute of Embedded Engineering of Korea (대한임베디드공학회)
권호 표지
DOI QR코드

DOI QR Code

Neural Network-based FMCW Radar System for Detecting a Drone

소형 무인 항공기 탐지를 위한 인공 신경망 기반 FMCW 레이다 시스템

  • Jang, Myeongjae (Korea Advanced Institute of Science and Technology) ;
  • Kim, Soontae (Korea Advanced Institute of Science and Technology)
  • Received : 2018.10.08
  • Accepted : 2018.11.16
  • Published : 2018.12.31
Download PDF
⟨ Previous Next ⟩

Abstract

Drone detection in FMCW radar system needs complex techniques because a drone beat frequency is highly dynamic and unpredictable. Therefore, the current static signal processing algorithms cannot show appropriate detection accuracy. With dynamic signal fluctuation and environmental clutters, it can fail to detect a drone or make false detection. It affects to the radar system integrity and safety. Constant false alarm rate (CFAR), one of famous static signal process algorithm is effective for static environment. But for drone detection, it shows low detection accuracy. In this paper, we suggest neural network based FMCW radar system for detecting a drone. We use recurrent neural network (RNN) because it is the effective neural network for signal processing. In our FMCW radar system, one transmitter emits FMCW signal and four-way fixed receivers detect reflected drone beat frequency. The coordinate of the drone can be calculated with four receivers information by triangulation. Therefore, RNN only learns and inferences reflected drone beat frequency. It helps higher learning and detection accuracy. With several drone flight experiments, RNN shows false detection rate and detection accuracy as 21.1% and 96.4%, respectively.

Keywords

OBDDBE_2018_v13n6_289_f0001.png 이미지

그림 1. FMCW 레이다 비트 주파수 Fig. 1 FMCW radar beat frequency

OBDDBE_2018_v13n6_289_f0002.png 이미지

그림 2. RNN 학습 및 추론 과정 Fig. 2 RNN learning and inferencing process

OBDDBE_2018_v13n6_289_f0003.png 이미지

그림 3. 시스템 하드웨어 구성 Fig. 3 System hardware

OBDDBE_2018_v13n6_289_f0004.png 이미지

그림 4. 비트 주파수 탐지 과정 Fig. 4 Beat frequency detection process

OBDDBE_2018_v13n6_289_f0005.png 이미지

그림 5. 비트 주파수 흐름 Fig. 5 Beat frequency stream

OBDDBE_2018_v13n6_289_f0006.png 이미지

그림 6. CFAR 기법과 RNN의 오탐지 발생률 (좌) 및 탐지율 (우) Fig. 6 False detection rate (Left) and detection accuracy (Right) of CFAR algorithm and RNN

표 1. FMCW 레이다 제원 Table 1. FMCW radar parameters

OBDDBE_2018_v13n6_289_t0001.png 이미지

표 2. 드론 사양 Table 2. Drone specifications

OBDDBE_2018_v13n6_289_t0002.png 이미지

표 3. 드론 비행 시나리오 Table 3. Drone flight scenario

OBDDBE_2018_v13n6_289_t0003.png 이미지

References

  1. H. Rohling, "Radar CFAR Thresholding in Clutter and Multiple Target Situations," Proceedings of IEEE Transactions on Aerospace and Electronic Systems, Vol. AES-19, No. 4, pp. 608-621, 1983. https://doi.org/10.1109/TAES.1983.309350
  2. C. Jeong, Y. Jung, S. Lee, "Neural Network-based Radar Signal Classification System Using Probability Moment and ApEn," Journal of Soft Computing, Vol. 22, No. 13, pp. 4205-4219, 2018. https://doi.org/10.1007/s00500-017-2711-7
  3. M. Jan, "Radar Signal Identification Using a Neural Network and Pattern Recognition Methods," Proceedings of IEEE Telecommunications and Computer Engineering, pp. 79-83, 2018.
  4. J.M. García, D. Zoeke, M. Vossiek, "MIMO-FMCW Radar-Based Parking Monitoring Application With a Modified Convolutional Neural Network With Spatial Priors," Proceedings of IEEE Access, Vol. 6, pp.41391-41398, 2018. https://doi.org/10.1109/ACCESS.2018.2857007
  5. R. Perez, F. Schubert, R. Rasshofer, E. Biebl, "Single-Frame Vulnerable Road Users Classification with a 77 GHz FMCW Radar Sensor and a Convolutional Neural Network," Proceedings of IEEE International Radar Symposium, pp. 1-10, 2018.
  6. B. Vandersmissen, N. Knudde, A. Jalavand, I. Couckuyt, A. Bourdoux, "Indoor Person Identification Using a Low-Power FMCW Radar," Proceedings of IEEE Transactions on Geoscience and Remote Sensing, Vol. 99, pp. 1-12, 2018.
  7. A. Krizhevsky, I. Sutskever, G.E. Hinton, "Imagenet Classification with Deep Convolutional Neural Networks," Proceedings of Advances in neural information processing systems, pp. 1097-1105, 2012.
  8. B. Kim, "(A) Study on Radar for Micro-drone Detection and Target Classification Based on Deep Learning," Ph.D. Dissertation, 2017.
  9. C. Szegedy, W. Liu, Y. Jia, P. Sermanet, S. Reed, D. Anguelov, A. Rabinovich, "Going Deeper with Convolutions," Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 1-9, 2015.
  10. E.D. Solovyeva, "Types of Recurrent Neural Networks for Nonlinear Dynamic System Modeling," SCM, 2017 XX IEEE International Conference, pp. 252-255, 2017.
  11. M. Kang, M. Lee, S. Kim, "Leakage and Clutter Suppression on FMCW Radar System for Small Unmanned Aerial Vehicle Detection," Proceedings of 2017 Defense SW/ICT Convergence Conference, SIGDS, 2017. (in Korean)
  12. C. Grimm, T. Breddermann, R. Farhoud, T. Fei, E. Warsitz, R. Haeb-Umbach, "Discrimination of Stationary from Moving Targets with Recurrent Neural Networks in Automotive Radar," Proceedings of IEEE MTT-S International Conference, pp. 1-4, 2018.
  13. D.A. Brooks, O. Schwander, F. Barbaresco, J.Y. Schneider, M. Cord, "Temporal Deep Learning for Drone Micro-Doppler Classification," Proceedings of IEEE International Radar Symposium, pp. 1-10 2018.
  14. M. Xu, C. Fan, J.D. Paden, G.C. Fox, D.J. Crandall, "Multi-Task Spatiotemporal Neural Networks for Structured Surface Reconstruction," arXiv, 1801.03986, 2018.
  15. M. Lee, S. Kim, "Performance Controllable Shared Cache Architecture for Multi-Core Soft Real-Time Systems," Proceedings of IEEE Computer Design, pp. 519-522, 2013.
  16. T. Park, S. Kim, "Dynamic Scheduling Algorithm and Its Schedulability Analysis for Certifiable Dual-criticality Systems," Proceedings of Embedded Sofrware, pp. 253-262, 2011.