Gunshot Technology – Hammers or Pliers? Depends on the task at hand
The science and technology of Gunfire detection has existed for nearly two decades. The basic technology is designed to detect the acoustic event (sound) created when a gun is fired. This is no easy feat as there are only two possible types of sounds that are created when a gun is fired. One is the “boom” of a muzzle blast and the other is the “snap” of a supersonic bullet (moving faster than the speed of sound).
As such Gunshot detection technologies have been fundamentally divided into two basic categories:
1. Supersonic crack detection
2. Muzzle blast detection
Supersonic Crack Detection
In the late ‘90s and early 2000s, a number of supersonic detection technologies were developed to detect supersonic bullets as they flew right past a sensor faster than the speed of sound. This military counter-sniper technology has been traditionally used for small areas with line-of-sight to assailants. These counter-sniper technologies have been used successfully in military deployments in many conflict zones. In such cases, the sensors providing the detection capability are deployed in the same place as (i.e. “collocated with”) the very thing that is the target of the sniper—e.g. a Humvee or an individual soldier. To put it another way, counter-sniper technology makes an a priori assumption that the counter-sniper sensor will be deployed in the line of fire. In addition the weapons used by insurgent snipers are typically capable of firing supersonic munitions. The advantage of this technology approach is that it detects very fast; produces very few false positives (there are not a lot of things that sound like a bullet breaking the sound barrier). The disadvantages are that it only detects supersonic weapons when the bullet passes directly by the sensor limiting its detection capability to very narrow use cases in battlefield deployments.
Muzzle Blast Detection
Muzzle blast detection is a non-line-of-sight, wide-area protection that works at long range. The technology generally falls into two categories:
• single point sensors – which operate in close proximity to the gunfire source
• distributed sensor arrays – which collaborate to produce detections
Both approaches detect the muzzle blast and have the ability to hear a distant gunfire boom or bang. In non-battlefield settings, muzzle blast approaches have the advantage of covering much larger areas with comparatively fewer sensors, as well as permitting bullets to be fired in any direction (at the sensor or not).
Single Point Sensors vs. Distributed Sensor Arrays
Single point sensors, sometimes called proximity sensors, generally require line-of-sight for accuracy and are commonly used to point individual video cameras located on top of, or right next to, sensors towards the origin. They are quite sensitive to indirect path sound (sound which has refracted or bent around, e.g. buildings), echoes, and multipath (multiple sounds produced by overlapping echoes), because they do not commonly have the capacity to differentiate the “correct” direction from a potentially incorrect direction.
Distributed Sensor Arrays (this is what we do).
Unlike counter-sniper systems and proximity sensors, distributed sensor array networks do not trigger because a single sensor hears a noise. Instead, they require multiple sensors to hear a noise over a short period of time (a few seconds) and in a pattern mathematically consistent with that sound having originated at a single location. There are two general approaches:
1. multi-lateration based on time difference of arrival (TDoA) and
2. triangulation (or generally multi-angulation) based on angle of arrival (AoA).
ShotSpotter technology uses a patented combination of both TDoA and AoA. (It also detects and uses supersonic bullet crack noises and adds them to the mathematical solution when it is deployed in configurations likely to be shot at.) To learn more about how ShotSpotter leverages the best of these gunshot detection technologies, read our whitepaper.
While many approaches have been used and tried, no two approaches are the same or capable of addressing the same use case or application. Users needing gunshot detection should have an objective in mind for their individual circumstance. For example, the needs and requirements for detecting gunfire in battlefield situations are drastically different from what is required in large urban settings. In a battlefield scenario, the interest is in understanding the direction from where the gun is firing in relation to the target. In a city situation, law enforcement is interested in knowing what type of gunfire is occurring at any location across the span of a city. Both of these scenarios require different scientific approaches and different solutions to be effective at detecting the gunfire.