Combined RF-Based Drone Detection and Classification

• DOI: 10.1109/tccn.2021.3099114
• Publication Date: 2021-07-26T21:02:46Z
• Authors: Sanjoy Basak; Sreeraj Rajendran; Sofie Pollin; Bart Scheers
• Keywords: Drone; SIGNAL (programming language); Detection theory

Despite several beneficial applications, unfortunately, drones are also being used for illicit activities such as drug trafficking, firearm smuggling or to impose threats security-sensitive places like airports and nuclear power plants. The existing drone localization neutralization technologies work on the assumption that has already been detected classified. Although we have observed a tremendous advancement in sensor industry this decade, there is no robust detection classification method proposed literature yet. This paper focuses radio frequency (RF) based using signature of transmitted signal. We created novel RF dataset commercial presented detailed comparison between two-stage combined framework. performance both frameworks single-signal simultaneous multi-signal scenario. With analysis, show You Only Look Once (YOLO) framework provides better compared Goodness-of-Fit (GoF) spectrum sensing scenario good comparable Deep Residual Neural Network (DRNN)