It’s so common to see video from military drones that nobody questions why the snippets are silent. Well, almost no one. A Dutch firm, maker of tiny, ultra-sophisticated microphones, is putting the devices on drone wings.
The object is not to record the devastation of attacks, but to stop attacks by simultaneously locating and even identifying multiple snipers and artillery crews for return fire.
Acoustic vector sensors, as the systems are called, can pinpoint, record and track sounds with the same accuracy as a radar can capture images of solid objects. But these airborne sensors can monitor a 360-degree sphere measured in kilometers simultaneously.
Hans-Elias de Bree, inventor of the sensors, said they operate at a small fraction of the cost of radar systems. And unlike radar installations, which announce themselves as a target by broadcasting energy beams, acoustic vector sensors are passive listeners, so they are harder to find and disable.
Better than today’s “web cams”
De Bree almost audibly sniffs at current drones.
“They are very nice Web cams,” he said. “They can tell you something about what they see when they are looking right at it.” De Bree is director of research and development for Netherlands-based Microflown Technologies.
Microflown, a privately held company, also makes sound-detecting devices used by automakers to make their vehicle interiors quiet. But much of the company’s growth and research is directed at the military applications. The new technology already has found buyers in the United States, Dutch and Indian armies, he said.
De Bree originally saw it as a tool for ground troops who are under attack and who urgently need to locate those firing on them, and as an affordable alternative for small-plane pilots to radar systems. It was only a matter of time before someone realized you could cross the two missions to take advantage of the rise of drones.
The principle underlying Microflown’s sensors is as straightforward as the technology is mind-bending.
The acoustic vector sensor is, basically, an anemometer, which measures wind speed. It hears sound throughout a sphere when aloft, even through the noise of an engine and the swirling, gusting winds through which an aircraft flies. Anything making noise disturbs that same air, and can be detected by a single sensor.
One sensor is sufficient because it is measuring the motion of air particles, not the motion of sound waves traveling through the air.
Multiple sensors—perhaps deployed on multiple aircraft and ground vehicles and carried by troops—can create a wider, more-detailed picture, but aren’t required as is necessary for triangulation.
The actual anemometer is the size of a small match head—one millimeter wide, two millimeters long and 300 micrometers thick. On an aircraft, it could be mounted flush on the fuselage’s skin or on a wingtip. The biggest danger for sensors is the gunk and water in the air.
Right now, protection consists of a bullet-shaped windshield with a collar of solid foam nickel that allows air particles to enter the sensor.
The challenges of signal processing
That’s the easy part.
Inside the sensor are two resistive platinum strips, each 200 nanometers thick by 10 micrometers wide, stretched parallel across a gap and heated to 392 degrees Fahrenheit while operating.
Air particles flowing through the gap cause the temperature of each strip to vary. This enables the sensor to count air particles, thereby measuring sound intensity. Just as important: The pattern of cooling between the two strips records particle movement, which reveals the x, y and z coordinates of where the sound originated.
The information is messaged by a pedestrian personal computer and analyzed by Microflown’s proprietary signal-processing software. The software removes ambient engine, air and ground noise.
Mechanical engineering experts say Microflown’s idea is solid, pausing only over the filtering, of all things. Technically, scrubbing sound of extraneous noise is called signal processing, and is a tricky task.
Once scrubbed, however, the system can create a log of sound signatures unique to each source. That means that after the first encounter, it can identify which rifle, mortar, howitzer or aircraft was being used.