Electro-Acoustic Path of the Detector for Detection of Small Unmanned Aerial Vehicles

Abstract

Small unmanned aerial vehicles have become increasingly popular in recent years for both commercial and recreational use. However, with the onset of war, the threat of using such drones for illegal purposes such as reconnaissance operations and terrorist attacks on the most vulnerable and important targets — military equipment, government buildings, airports or places with a large concentration of people — has rapidly increased. As a result, many methods have been developed for detecting drones, but the most promising of these is considered to be the acoustic method. This article discusses the implementation of a simple and inexpensive acoustic detector with a single microphone, enabling the detection of the sound emitted by the drone engines and propellers, which can then be analyzed to determine the location and trajectory of the unmanned aircraft. To register sound waves, it was proposed to use the electret microphone CMA-4544PF-W with a foam windscreen, and to ensure a wide dynamic range and protect the detector from overload, an amplifier with an automatic gain control system with feedback regulation based on the integrated circuit MAX9814. For further digital signal processing on a PC, an analog-to-digital converter with a sampling frequency of 48 kHz and a bit depth of 16 bits will be used, as well as an external sound card CM6206 with a line input. As a result of test flights, a database of audio files of the noise of the DJI Mavic 2 Pro quadcopter was created, which is now actively used by the military for reconnaissance, the accuracy of object detection by the acoustic detector was studied. The distance from the drone to the microphone affected the accuracy of detection, and the maximum value of reliable detection was 40 meters. The device successfully performs its functions, given its low cost and ease of use, but requires improvement to enhance its characteristics. The spectra of the obtained audio recordings of quadcopter emissions make it possible to determine the main frequencies of tones, the number of which coincides with the number of electric motors and are important features for further identification of the drone.