Acoustic detection results for a small unmanned aircraft system extrapolated over range
Geoffrey H. Goldman · The Journal of the Acoustical Society of America · 2016
Small unmanned aircraft systems (SUAS) are becoming cheaper with more advanced reconnaissance and surveillance capabilities. Low-cost sensors and countermunitions are needed to defeat this asymmetric threat. The first step is to detect SUAS using low-cost sensors such as an array of microphones with robust signal processing algorithms. An analysis of six detection algorithms was performed using acoustic data generated by a class I SUAS at ranges of 50 to 650 m and measured with a small tetrahedral microphone array with 20-cm length arms. The detection algorithms were based upon the peak power in delay-and-sum, filtered delay-and-sum, and adaptive delay-and-sum beamforming algorithms. To test the performance of the algorithms at longer ranges, the measured signal, modeled as target plus additive noise, was modified. First, the signal from the target was attenuated using Bass’s model and spherical attenuation, and then, the noise was adjusted to maintain an unbiased estimate of the measured power spectrum density function. Receiver operation characteristics (ROC) curves were generated and the performance of the algorithms as a function of range was evaluated.