RF-Enabled Deep-Learning-Assisted Drone Detection and Identification: An End-to-End Approach is a study by Syed Samiul Alam, Md Habibur Rahman, Raihan Bin Mofidul, Md Morshed Alam, Ida Bagus Krishna Yoga Utama, e Yeong Min Jang.

This article addresses the rising security concerns related to unmanned aerial vehicles (UAVs) and proposes an end-to-end deep-learning-based model for their detection and identification based on radio frequency (RF) signatures. The model utilizes multiscale feature-extraction techniques without manual intervention, ensuring efficient extraction of enriched features and better generalization capability with lower computational time.

Residual blocks are incorporated to handle complex representations and address vanishing gradient issues during training. The model demonstrates effectiveness in the presence of interfering signals such as Bluetooth and WIFI. Evaluation across various signal-to-noise ratios (SNR) yields impressive results, with an overall accuracy of 97.53%, precision of 98.06%, sensitivity of 98.00%, and F1-score of 98.00% for RF signal detection in the CardRF dataset. The proposed model’s inference time of 0.37 milliseconds outperforms existing work, making it a promising solution for real-time UAV detection and identification in surveillance systems.

Data di pubblicazione– April 2023

RF-Enabled Deep-Learning-Assisted Drone Detection and Identification: An End-to-End Approach contains the following major sections:

  • Introduzione
  • Metodologia
  • Experimental Results
  • Conclusioni

ll articles published by MDPI are immediately available worldwide under an open-access license. No special permission is required to reuse all or part of the article published by MDPI, including figures and tables. Any part of the article may be reused without the consent for articles published under an open-access Creative Common CC BY license, provided the original article is clearly cited. For more information, please refer to https://www.mdpi.com/openaccess.

C-UAS Hub non è proprietaria di questi contenuti e fornisce agli utenti un link al sito fondo della pagina per accedervi nella sua posizione originale. Ciò consente agli autori di tenere traccia di importanti metriche relative al loro lavoro. Il merito va al legittimo proprietario.

Autori-Syed Samiul Alam, Md Habibur Rahman, Raihan Bin Mofidul, Md Morshed Alam, Ida Bagus Krishna Yoga Utama, and Yeong Min Jang

Vedi anche...

Radar bistatico passivo: Rilevamento del bersaglio e interferenze

Rilevamento e tracciamento dei droni grazie ai segnali di identificazione RF

Apprendimento profondo su dati multisensore per applicazioni contro gli UAV

Crediti per l'immagine: envatoelements by atercorv