Acoustic-Based UAV Detection Using Late Fusion of Deep Neural Networks

  1. Casabianca, Pietro
  2. Zhang, Yu
  3. Diego González Aguilera 1
  1. 1 Universidad de Salamanca
    info

    Universidad de Salamanca

    Salamanca, España

    ROR https://ror.org/02f40zc51

Revista:
Drones

ISSN: 2504-446X

Año de publicación: 2021

Volumen: 5

Número: 3

Páginas: 54

Tipo: Artículo

DOI: 10.3390/DRONES5030054 GOOGLE SCHOLAR lock_openAcceso abierto editor

Otras publicaciones en: Drones

Resumen

Multirotor UAVs have become ubiquitous in commercial and public use. As they become more affordable and more available, the associated security risks further increase, especially in relation to airspace breaches and the danger of drone-to-aircraft collisions. Thus, robust systems must be set in place to detect and deal with hostile drones. This paper investigates the use of deep learning methods to detect UAVs using acoustic signals. Deep neural network models are trained with mel-spectrograms as inputs. In this case, Convolutional Neural Networks (CNNs) are shown to be the better performing network, compared with Recurrent Neural Networks (RNNs) and Convolutional Recurrent Neural Networks (CRNNs). Furthermore, late fusion methods have been evaluated using an ensemble of deep neural networks, where the weighted soft voting mechanism has achieved the highest average accuracy of 94.7%, which has outperformed the solo models. In future work, the developed late fusion technique could be utilized with radar and visual methods to further improve the UAV detection performance.

Referencias bibliográficas

  • Choi-Fitzpatrick, (2014), J. Int. Aff., 68, pp. 19
  • DJIhttps://www.dji.com/uk/mini-2?from=store-product-page
  • 10.2514/6.2020-0491
  • 10.1117/12.2533039
  • 10.2352/ISSN.2470-1173.2017.10.IMAWM-168
  • 10.1109/TVT.2019.2893615
  • 10.1155/2008/834973
  • 10.1155/2015/209814
  • 10.1109/SACI.2016.7507430
  • 10.1109/IWCMC.2019.8766732
  • 10.5446/44845
  • 10.1115/GT2018-75286
  • 10.1109/JSTSP.2019.2908700
  • 10.23919/EUSIPCO.2018.8553106
  • 10.1109/TII.2020.3007152
  • 10.1109/TII.2020.3005965
  • 10.1007/s10994-006-9019-7
  • 10.1016/j.ecoinf.2019.05.007
  • Scikit-Learnhttps://scikit-learn.org/stable/about.html
  • 10.25080/majora-7b98e3ed-003
  • 10.1007/978-981-10-6898-0_19
  • Masters, (2018), arXiv
  • Goodfellow, (2016)