Acoustic detection of low flying aircraft

A passive acoustic system, consisting of a set of acoustic nodes and a central processing node, is a good candidate for the covert detection of low flying aircraft as they cross our nation's borders. Small arrays of microphones have been used to estimate aircraft speed, heading, frequency spectrum and altitude at the closest point of approach. We extend this work by considering maneuvering aircraft and by testing a narrowband algorithm that gives angle estimates sooner, and over a longer period of time, than the broadband version. A new narrowband algorithm is provided for estimating position versus time as the aircraft flies through a grid of widely spaced omni-directional microphones (separated by several miles). We classify aircraft as helicopter, piston engine aircraft (specifically, numbers of propeller blades and cylinders), turboprop aircraft or jet aircraft. Our approach has been validated via testing at various airports against ground truth data.