Intrusion Detection Framework for Invasive FPV Drones Using Video Streaming Characteristics

Cheap commercial of-the-shelf (COTS) First-Person View (FPV) drones have become widely available for consumers inrecent years. Unfortunately, they also provide low-cost attack opportunities to malicious users. Thus, efective methods to detect the presence of unknown and non-cooperating drones within a restricted area are highly demanded. Approaches based on detection of drones based on emitted video stream have been proposed, but were not yet shown to work against other similar benign traic, such as that generated by wireless security cameras. Most importantly, these approaches were not studied in the context of detecting new unproiled drone types. In this work, we propose a novel drone detection framework, which leverages speciic patterns in video traic transmitted by drones. The patterns consist of repetitive synchronization packets (we call pivots), which we use as features for a machine learning classiier. We show that our framework can achieve up to 99% in detection accuracy over an encrypted WiFi channel using only 170 packets originated from the drone within 820ms time period. Our framework is able to identify drone transmissions even among very similar WiFi transmission (such as video streams originated from security cameras) as well as in noisy scenarios with background traic. Furthermore, the design of our pivot features enables the classiier to detect unproiled drones in which the classiier has never trained on and is reined using a novel feature selection strategy that selects the features that have the discriminative power of detecting new unprofiled drones.