Proceedings of the 13th EAI International Conference on Mobile Multimedia Communications, Mobimedia 2020, 27-28 August 2020, Cyberspace

Research Article

Radio Frequency Fingerprinting Driven Drone Identification Based on Complex-valued CNN

Download310 downloads
  • @INPROCEEDINGS{10.4108/eai.27-8-2020.2295045,
        author={Hao  Gu and yu  wang and Guan  Gui and Sheng  Hong and hao  huang and Jie  Yang and Miao  Liu and Jinlong  Sun and yun  lin},
        title={Radio Frequency Fingerprinting Driven Drone  Identification Based on Complex-valued CNN },
        proceedings={Proceedings of the 13th EAI International Conference on Mobile Multimedia Communications, Mobimedia 2020, 27-28 August 2020, Cyberspace},
        publisher={EAI},
        proceedings_a={MOBIMEDIA},
        year={2020},
        month={11},
        keywords={drone identification complex-valued cnn intelligent recognition rf fingerprinting deep learning},
        doi={10.4108/eai.27-8-2020.2295045}
    }
    
  • Hao Gu
    yu wang
    Guan Gui
    Sheng Hong
    hao huang
    Jie Yang
    Miao Liu
    Jinlong Sun
    yun lin
    Year: 2020
    Radio Frequency Fingerprinting Driven Drone Identification Based on Complex-valued CNN
    MOBIMEDIA
    EAI
    DOI: 10.4108/eai.27-8-2020.2295045
Hao Gu1, yu wang1, Guan Gui1,*, Sheng Hong1, hao huang2, Jie Yang1, Miao Liu1, Jinlong Sun3, yun lin4
  • 1: Nanjing University of Posts and Telecommunications
  • 2: Nanjing University of Posts and Telecommunications,Nanjing,Jiangsu,China
  • 3: Nanjing University of Posts and Telecommunications, china
  • 4: Harbin Engineering University
*Contact email: guiguan@njupt.edu.cn

Abstract

Drone detection and identification technique is of great significance both in the military and civilian fields. Radio frequency (RF) fingerprinting of drone is considered as one of promising techniques due to its uniqueness. Deep learning based RF fingerprinting identification technique can extract hidden features in RF data and then achieve excellent performance. Motivated by this idea, this paper proposes a drone identification method using complex-valued convolutional neural network (CNN) algorithm with higher classification accuracy and faster equipment running time. The complex-valued CNN method convolves the complex convolutional kernel and the real and imaginary parts of the data features separately. In order to verify the proposed method, five state-of-the-art recognition algorithms are adopted to compare their recognition performance and equipment efficiency. Simulation results show that our proposed drone identification method can efficiently recognize the signal of various drones within less computation time.