Use of drones, or unmanned aerial vehicles (UAVs) are becoming more prevalent. As this technology improves, security threats from adversary drones are rapidly evolving including spying on sensitive facilities and the delivery of contraband materials, such as drugs to prisons. Locating these devices is critical, but different geographic and electromagnetic environments often require very different approaches to detection, and active methods like radar are not always viable. To counter this, passive techniques, like acoustic monitoring, must be developed.
Like helicopters, drones typically rely on a propeller system to remain aloft, and these multirotor devices produce a constant noise source throughout flight.
We developed a 64-element microphone array for use in sound-source location. Applying this to a UAV in a flight trial, we were able to generate an image which is shown left. It illustrates sound pressure in dB, and the red cross dictates the maximum area of pressure in the scanning grid. Unfortunately, as beamforming algorithms locate any sound source, there was a risk of drones being overlooked in favour of other sources such as voices or wind noise.
An algorithm has been written which can identify the characteristics of the sound produced by multirotor drones. This significantly reduces the likelihood of incorrectly identifying other noise sources as drones.
This approach allows the system to hone in on the noise generated by the propellers, and exclude external sound sources. This provides an unobtrusive detection capability, able to detect UAVs without specific permits. The system is completely passive, operates without interfering with other radar systems, does not require a broadcast licence, and most importantly, in a military setting, would not alert adversaries to your location.
Related news, insights and innovations
Find out more about our cutting-edge expertise.