AcousticPrint
Harini Kolamunna, Junye Li, Thilini Dahanayaka, Suranga Seneviratne, Kanchana Thilakaratne, Albert Y. Zomaya, Aruna Prasad Seneviratne · 2020
Malicious or improper use of drones can pose significant privacy and security threats in both civilian and military settings. There are many situations where it requires to detect the presence of a drone and identify the exact model to be used in applications such as law enforcement depending on the size and capabilities of different models. Nonetheless, this remains a challenging task, especially in low visibility, limited access, or hostile environments. In this paper, we propose to use acoustic signatures to identify the make and the model of drones. We achieved 94% accuracy in a closed set scenario and 80% accuracy in a more challenging open set scenario.