Software Defined Radio and Wireless Acoustic Networking for Amateur Drone Surveillance

Abstract

The advancement of the micro-electro-mechanical sensory industry and open source autopilot stacks, have dramatically reduced the cost and difficulty of making highly maneuverable UAVs. Such ease in flying drones has caused concerns about privacy, public safety, and security. One of the major threats is inadequate control of flying small drones over sensitive areas. In this research, we address this problem through a dual approach: detection and eviction. We propose a distributed system to identify the appearance and the approximate position of unwelcome drones using a wireless acoustic sensor network and machine learning algorithms. Next, we integrate software defined radio transceivers along with machine learning algorithm into a framework to specify and decode the UAV's telemetry protocols. Using decoded information, we can then send control commands to evict the trespassing drone away before sending aggressive surveillance UAVs.