Acoustic Signature: Measuring What Gives Your Platform Away

Detection Moved to Passive

Counter-UAS detection has shifted decisively toward passive methods, for the straightforward reason that RF detection fails against platforms that do not radiate — fiber-guided aircraft, or terminal-phase platforms that shut down their video transmitter on final. Fielded acoustic detectors using four-channel MEMS arrays now achieve 100-300 m detection and bearing on multirotors, with the range depending heavily on ambient noise.

Meanwhile, acoustic signature remains almost entirely unmeasured on the development side. Programmes that will spend months on radar cross-section or thermal signature typically have no acoustic data whatsoever, and cannot answer basic questions about how far away their aircraft is audible.

What Actually Radiates

Multirotor acoustic output is dominated by blade passage frequency: BPF = rotor RPM × blade count / 60. A 6,000 RPM two-blade rotor produces a 200 Hz fundamental plus harmonics. This tonal structure is what makes acoustic classification effective — it is a distinctive, periodic signature that separates cleanly from broadband environmental noise, and it is precisely what detection algorithms are built to find.

Three factors drive how far it carries. Tip speed above all: acoustic power scales roughly with the fifth to sixth power of tip speed, so a modest reduction produces a large signature reduction. Blade count and spacing, because uneven spacing spreads energy across frequencies and reduces the tonal peak that classifiers key on. And motor and ESC switching noise, which contributes high-frequency content that attenuates faster with distance but is highly distinctive close in.

Field Example: Larger Rotors, Quieter Aircraft

On a reconnaissance quadrotor where acoustic detection range was an explicit requirement, we measured signature across the throttle range with a calibrated array at fixed distances.

The aircraft produced a 148 Hz fundamental at hover with strong second and third harmonics. Measured detection range against a representative MEMS array was 240 m in 35 dBA ambient. The team's proposed fix was acoustic damping on the arms, which addressed structure-borne noise contributing very little to the far field.

The effective change was aerodynamic: larger-diameter rotors turning slower for the same thrust. Moving from 15-inch rotors at 5,900 RPM to 18-inch at 4,400 RPM dropped tip speed by roughly 20% and measured far-field SPL by 7 dB, taking detection range to 135 m. Hover endurance improved 6% as a secondary benefit, because larger slower rotors are more efficient in hover. The signature requirement and the endurance requirement pointed the same direction, which is not always the case but is worth checking before assuming a trade.

Measuring It Without a Laboratory

You do not need an anechoic chamber. A calibrated measurement microphone, an open site with low ambient noise, and a disciplined procedure produce usable comparative data. Measure at fixed ground distances with the aircraft at fixed altitude, record ambient before and after every run, and capture the full throttle range rather than hover alone — climb is substantially louder and is often when the aircraft is first detected.

Report the spectrum, not just an A-weighted level. A single dBA figure tells you nothing about whether a classifier will identify the platform, and classification range is frequently what matters operationally rather than raw detection.