Your browser version is outdated. We recommend that you update your browser to the latest version.

SIGN UP to our free eNewsletter

Fields marked with * are required.

DEFSECmedia

DEFSEC Media is New Zealand's defence, security and fire B2B/B2G publishing group. Our leading magazines, Line of DefenceNZ Security and Fire NZ are read by key business, government and military decision makers and influencers. This website is the online home of cutting-edge content from each of our titles.

Innovation Central

Automation & Electronics

Itron

Sky wars: Understanding the dark art of drone detection

FEATURES: Line of Defence, April 2016

The Blighter AUDS system combines electronic scanning radar target detection, electro-optic tracking/classification and directional RF inhibition capabilityThe Blighter AUDS system combines electronic scanning radar target detection, electro-optic tracking/classification and directional RF inhibition capability

 

The rapid development of the drone, or UAV (Unmanned Aerial Vehicle) industry has rattled nerves within government, military and commercial contexts. Concerns about malevolent actors using UAVs in airport, prison, military base, battlespace, national border and critical national infrastructure scenarios have prompted the creation of systems aimed at detecting, disabling and/or destroying them.

It’s a dynamic game. As military developed drones have paved the way for the harnessing of drone technology by nefarious actors, this in turn has fuelled the race to operationalize drone detection, and companies from Airbus to a plethora of start-ups are getting in on the act. Ironically, the anti-drone rush is also leading to the development of anti-detection responses.

According to Rachel Stohl of the Stimson Center in a recent Cipher Brief interview, “It is sort of a chicken and the egg thing. You create a technology, and then you create something to counter that technology, and then you create a new technology or a new aspect of that technology, or a new system that is associated with the technology. And then more countermeasures are built.”

Too close for comfort

In 2014, pilots reported to the US Federal Aviation Administration 238 instances of drones flying in the vicinity of an airplane approaching an airport. In June and July of 2015, 275 such reports were filed. As of the end of February, the FAA is receiving at least 100 reports each month.

Last July, a drone delivered an illicit package of heroin, marijuana and tobacco to the yard of the Mansfield Correctional Institution in Ohio while inmates were outside, sparking a fight. This incident, stated the Daily Telegraph (UK), created a new front in the war on drugs in prisons.

In February, the Seattle Times reported that a UAV flew over US Naval Base Kitsap-Bangor in Silverdale, USA. Navy investigators want to know who was flying it and why, and since nuclear submarines are based at Kitsap-Bangor, this particular incident is being taken very seriously.

Such aerial incursions are becoming increasingly commonplace, and against the prospect of increased – and weaponised – use by criminal and terrorist actors in not only domestic but also military deployment scenarios, developers, regulators and strategists are playing catch-up.  Welcome to the age of close-quarter unmanned mechanised battle.

Drone defence options

The size and agility of drones make them difficult for conventional radar systems to spot, precisely because manufacturers have traditionally employed software to prevent birds from registering and crowding radar screens. Engineers are now having to develop algorithms to distinguish drones from birds.

Other options have been considered, including ‘geo-fencing’, which involves coding a drone’s firmware to prevent it from flying into designated areas, such as airports or government buildings. This, though, can be reprogrammed. Ultimately, such drone detection solutions are expensive, and potentially prone to compromise.

And then there’s the challenge of neutralising a drone once detected, for which the market is offering a range of options. Anti-UAV Defence System (AUDS) from Blighter Surveillance Systems uses radio beams to freeze drones in mid-air. Their latest version uses a quad-band radio frequency (RF) jammer that can disrupt all commercial drone licensed telemetry bands, and is effective against micro UAVs up to two kilometers away.

British engineering firm OpenWorks' solution uses a shoulder-mounted launcher that captures a transgressor UAV in a net from up to 100 metres away. The SkyWall Launcher system uses a compressed gas-powered launcher to fire a programmable projectile that deploys a large net designed to drape over the drone and take it down safely.

Another system, SkyJack, developed by Samy Kamkar, is a drone designed to autonomously seek out the wireless signals of other UAVs and take control of them while in flight. The system’s wireless adapters, once deployed, detect any nearby wireless connections in range and identifies those associated with other UAVs. It then disconnects these drones from their owners through raw packet injection, and takes over, turning them into zombies.

In an interesting mix of old and new technology, one company is training eagles to intercept and bring down drones from where they’re hovering for the Dutch Police.

But before a drone can be neutralised, it must first be successfully detected. According to Ross Taylor, UAV specialist at Auckland-based Tactical Solutions, the five drone detection technologies currently in most common use are audio, video, thermal, radar and radio frequency (RF).  “With the exception of RF detection,” says Taylor, each method is rather weak when used individually for drone target identification. They are very commonly paired with one another to provide better results.”

Audio

Audio detection utilises microphones which are programmed to alarm when they identify the specific sound signature produced by a drone’s propellers. Audio sensors are extremely prone to background ambient noise, and quickly become near useless in urban environments with plenty of vehicle and building noise. Wind and rain can also produce considerable noise which renders the systems useless.

Audio detection can also be relatively easy to defeat by simply altering the sound signature produced by the drone. This can be done by altering the shape or position of the propellers.

Audio detection systems have a very short effective range, most working at only 60 meters or less. This is because small drone propellers produce a high pitched noise that does not propagate long distances. A drone may not be audible – depending on size – until it is within 100 meters, which allows insufficient reaction time against a drone moving at 16 meters per second.

Audio detection with an effective range of 60 meters requires multiple units to be installed to secure even the smallest of locations, and for this reason audio remains the least effective method of drone detection.

Video

Video detection for drones is very software reliant, and only works out to a short distance. While zoom is possible to increase that distance, the image quality takes a major step backwards and lowers reliability. Most systems are effective out to 100 meters, but the higher the cost of the camera the further its range can be extended.

Cloud coverage or heavy rain can, however, reduce visibility, rendering the detection system useless. But it’s also worth noting that whenever a system is rendered ineffective due to inclement weather, those conditions also prohibit most consumer drones from flying.

Even in ideal weather conditions it is difficult to program video-only systems to properly identify a drone. A bird flying past the camera, for example, can trigger a false alarm given that most video surveillance systems do not distinguish between sources of movement. This becomes very counterproductive in heavy bird flight zones.

Multiple static cameras are required in order to cover a large area (given drone movement detection requires the detector to be stationary). Multiple cameras may even need to be used for a small area because of their limited fields of view. Multiple units mean higher costs and error rates.

As part of a multi-method detection solution, however, cameras can work well. Airbus Defense & Space GmbH radar, for example, can identify a small unmanned aircraft as far as 2.8 miles (4.5 km) away. Video cameras take over at closer range, able to identify a small drone from 1.5 km away.

Thermal and electro-optical

Thermal detection uses a thermal camera to measure the natural infrared radiation from an object and renders a viewable image. The problem is that consumer drones – with their plastic bodies and small electric motors – produce very little heat. In a hot climate it can become difficult to see them. Conversely, they can excel in penetrating inclement weather, and they are especially effective in cold environments where mechanical heat stands out more obviously.

Thermal detection is similar to video in its pitfalls, being software reliant and unable to organically identify specifically what a target actually is. Databases are constantly being updated to improve identification, but as new drones are constantly being rolled out with new heat signatures it’s an uphill battle.

Thermal detection does, however, excel at identifying gas powered drones, which are generally larger and produce a lot more heat than their electrically powered counterparts. The only issue with this is that gas powered drones are not widely available and are therefore less likely a threat compared to the ubiquitous plastic consumer drone.

Thermal is nevertheless one of the best complimentary detection methods. With another system providing information on where to point the camera and search, the operator is able to identify a drone target without relying unilaterally on a thermal video feed.

Radar

Radar, used predominantly in the detection of aircraft and ships, has been adapted to suit detection of much smaller objects. When detecting large objects such as an aircraft, shape and definition is distinguishable, however, clearly identifying the target becomes much more difficult when it comes to detecting a drone. This lack of clarity means that radar is almost always paired with another detection method.

Pairing radar with a visual detection method, such as video or thermal cameras (or even binoculars), creates an extremely robust system with a strong threat identification range. The radar will detect an incoming target at a long range and can then automatically align a camera for an operator to view and determine if the target is a threat.

Some drone detection systems that use radar and thermal cameras have had promising results at a range of 1,000 meters and above. These systems can be very flexible and also operated for the identification of other threats, such as incoming vehicles or personnel.

The Kelvin Hughes SharpEye SxV radar – enhanced for drone detection – is a good example of this. The system can be deployed as a mobile radar or as part of a multi radar and electro-optic camera system for mobile and semi-permanent requirements suitable for forward deployment scenarios.

Radio Frequency (RF)

Radio Frequency is the only identification method that excels at target identification on its own, especially in relation to off-the-shelf consumer drones. Designed to specifically identify the connection between a pilot’s controllers and their drone, it is possible for RF detectors to provide GPS co-ordinates of both the drone and its pilot if the model of drone is loaded into the detector’s database (which are constantly being updated).

Being able to identify the pilot’s position is one of RF’s most attractive attributes, as it then allows for the pilot to be located and managed.

RF detection has a very long range, with distances of 2,000 meters possible in ideal conditions. It is also possible for an RF detector to find drones operating under water and on the ground provided the RF signals are strong enough.

With such an amazing range and the potential for GPS location, RF is a clear front runner as an early warning consumer drone detection system. With one unit being able to cover an area that would otherwise take numerous cameras or audio systems to cover, and with a range that provides enough warning to initiate a response, RF is heads above.

CACI International’s RF-based SkyTracker, for example, can not only monitor drones near sensitive sites, it can also home in on the person flying the device by monitoring radio signals, giving the operator the ability to locate the perpetrator. In some cases, it may even be possible to get identifying information such as a device’s serial number.

The drone detection space is changing just as rapidly as the drone development space. As manufacturers jump in their droves onto the juggernaut that is the international UAV market, others are trying their hand at drone detection. As we look to the skies, the threats of the future may well not be the biggest or the fastest… they may be slow, very slow, and small, so small in some cases that they may be as difficult to locate and swat as a summer fly.

Back to Defence

Follow us...

We recommend on YouTube...