(DTD0003)
Written evidence submitted by Manchester Metropolitan University
Walker-Roberts, S & Hammoudeh, M
The drone industry is projected to be worth $17 billion by the end of 2024 [1] with over 100,000 shipments in 2016. It is therefore a viable proposition that the growth in drone uptake and use both commercially and by hobby pilots has been explosive. The media has not helped the case for legitimate uses of drones by showcasing highly targeted conditions whereby drones have been demonstrated to destroy airframes for heavy passenger aircraft though in reality conditions would have to be perfect for such precise structural damage to occur.
The present technological climate cannot cope with the increasing numbers of drones and the potential level of threat of malicious use of drones. Whilst there are few, if any, recorded cases in the UK of drones being used maliciously in any major incident other that the recent Gatwick disruption, the capability of malicious actors to commit atrocities is a fairly open one. Our present ability to detect and neutralise malicious drones in particular is poor and the investment required to improve that situation is not apparent.
CNI is “Critical National Infrastructure” as defined by the Centre for Critical National Infrastructure
There is no clear framework for CNI stakeholders to decide whether a drone is a threat, and if they determine drones to be as such, there is no legislation in place which equips stakeholders to deal with that potential threat. There is no consensus as to whom decides when a drone is a threat and any existing partnerships in regards to drones are formed from initiative and not as part of a national model or framework, therefore there is no uniform strategy to dealing with potentially rogue drones.
Drones are technologically limited as to the weight that they can carry and in general are not capable of highly sophisticated attacks by terrorists unless assisted by a very capable actor, probably a nation state.
The capability of countermeasures to protect CNI against potentially malicious drones is not commercially proven. In addition, the second and third order effects of available solutions are not tested or fully understood which poses serious challenges.
Electronic conspicuity refers to the ability to identify an aircraft.
The future of drones is likely to be strongly focussed on electronic conspicuity [2] whereby drones use air traffic control services to achieve safe aviation in controlled airspace for legitimate purposes such as blue-light drones in medical services and to swiftly identify anomalies in controlled airspace in conjunction with a shared intelligence network.
In general, there are far fewer malicious actors than legitimate ones. Drones can be of immense benefit to society if a safe scheme for their use can be established in controlled airspace. The likelihood of a drone being used for a malicious purpose is no greater than the existing chance of a malicious act taking place. With the correct systems and processes in place, drones stand to be an extremely safe technology in general.
No. The current technologies available in the UK to counter malicious drone activity is very limited and ineffective. Following the recent Gatwick drones incident, the police have warned against shooting drones down because of the risk that stray bullets may harm people and property. Key CNI assets have also demonstrated to us the potential danger to CNI assets themselves in using force to address a potentially malicious drone. Shooting explosive, chemical or nuclear armed drones over inhabited areas may lead to catastrophic results and is obviously not practical. The use of jammers and other equipment is also likely to be impractical in the aviation sector in particular, and therefore aerodrome CNI assets face even further challenges in removing the control of a malicious actor from a malicious drone and safely neutralising its threat.
Safer alternatives to capturing drones can be through using existing products such as SkyWall [3], which is designed by a British company. It is a shoulder mounted compressed air launcher to capture a drone. The operator of SkyWall targets the drone by firing shells containing nets to tangle the drone and bring it down to the ground safely using a parachute. Besides the challenge of identifying hostile airborne surveillance or potentially malicious activities in a restricted airspace and the need for a trained operator, the maximum range of the fire launcher is 100m only. This range is not enough to neutralise an armed drone safely.
These are various radio communication technologies, most are commonly used in mobile devices generally.
A widely available technology to counter drone activity is RF jamming devices that can shoot down high altitude unmanned aerial vehicles. Such devices have proven effective to respond to terrorist drones attacks or defend against drone assassinations. Jamming devices can effectively block GPS, BeiDou and GLONASS as well as CDMA, GSM, DCS, PCS, 3G, WiFi, VHF, UHF, 4G and LTE signals. Jammers readily available on the market have a maximum range of jamming of up to 2km coverage in a wide range of environmental conditions. Common anti-drone radio wave guns, e.g., the US made DroneDefenders [4], target the drone by a built-in sight and can land the drone only up to 400m away. Such guns are mobile, harmless to non-hostile devices, secure and work on radio control frequency disruption technology in a supervised manner. However, jamming a drone signal requires identifying the frequency a drone is operating on which can take time. Also, many drones have inbuilt ‘return home’ functions which are automatically started when the signal is lost.
Despite enormous security budgets to protect CNI assets such as airports, budgets for the police and for military intervention, the government engagement as well as the reasonable effectiveness and wide availability of jamming and capturing technologies are generally quite poor. The recent incident at Gatwick airport led to 120,000 passengers having their flights delayed or cancelled. Surprisingly, no drone(s) were captured, disabled or destroyed. Since then, Gatwick airport has announced a £5 million spending on counter-drone technology [5]; more than double the recently announced UK Government £2 million competition to counter the future threat of drones. For security reasons the details of the new technologies deployed at the airport is undeclared. The amount the Government is prepared to fund on individual projects by businesses and academic institutions is strikingly low and does not reflect the current hunger for technologies to address what is a high level of risk from malicious actors.
The future of signal jamming as an effective tool for disabling attack drones is questionable due to advances in jamming resistant technologies. For instance the Russian defence ministry's research foundation, the FPI, has a program to develop GPS-free drones [6]. Similar programs exists in many other countries, for example, the US Army launched a $59 million programme to develop GPS-independent drones [7]. The academic literature is rich in jammer localisation techniques that have been developed to estimate the position of a radio receiver by exploiting signal propagation phenomena, e.g., time of flight, angle of arrival and delay spread, to infer the distance to the radio transmitter adopting well-known propagation models. Other algorithms can also be utilised such as triangulation, angulation, trilateration, multilateration, and alteration. Applying these techniques to drones is predominantly practical and achievable depending on the information obtained from the signal level such as the Received Signal Strength (RSS).
Many techniques have been developed to defend against radio jamming attacks by malicious jammers. Most techniques use probabilistic methods and protocols to evade the jammed frequencies. Early techniques commonly focused on narrow band jamming. For instance, the singular-value decomposition (SVD)-based method[8] and the transversal filters [9] were designed to mitigate single-tone attacks. In malicious attack scenarios, the use of channel codes such as Bose-Chaudhuri-Hocquenghem (BCH) codes or convolutional codes, by drones has been proven to be highly effective even without knowing the jamming frequency. A number of adaptive techniques exist to mitigate jamming of GPS satellites, e.g., frequency/time domain filtering [10] and adaptive antenna array [11]. Another method of anti-jamming techniques is the utilisation of spread-spectrum, e.g., frequency hopping [12], to evade the jamming signal. Drones can simply execute a spatial evacuation approach to retreat from the jamming area when they experience communication failures. The research presented in [13] designs a cooperative spatial retreat solution to enable drones to cooperate for defeat the actions of the jammer.
The view from CNI assets was that there are many suggestions for countermeasures and these are well-documented and theorised but CNI assets were keen to point out that there are few commercially proven solutions which are effective for the needs of CNI. Furthermore, there was clear concern that little is understood about second order and third order effects for countermeasures in order to decide on the best procedures and deployment methods.
Regardless of the technique used to neutralise a drone, the first step in a successful defence solution is an early detection and tracking function. A drone detector can be a simple electronic device that recognises radio signals of electronic drones. Detectors are equipped with radio frequency sensors, acoustic sensors and cameras. Such devices can be integrated with drone defence systems to offer a fully autonomous cost effective threat elimination solution using image recognition and localisation algorithms or even artificial intelligence to differentiate hostile objects. Some drone detectors are capable of identifying the type of the drone and can provide an early warning on the presence of drones and send alerts to system owners which increases the time to handle an intrusion safely. To the best of our knowledge, despite the presence of commercial drone detector devices such as Dedrone, there is no or few anti-drone systems in critical government/military/industrial installations (e.g., prisons and airports) that integrate autonomous drone detection devices. Large premises such as a football stadium require more than 10 devices to achieve realistic detection accuracy, which poses a scalability challenge and makes traditional radar solutions more effective. Solutions such as AUDS (Anti-UAV Defence System) integrate electronic-scanning radar target detection, electro-optical (EO) tracking/classification and directional RF inhibition capability. AUDS radar-based drone detector (at distances of up to 10km), video track, target classification and directional RF inhibition capability provide maximum protection.
Consulting with CNI assets and aviation professionals, we understand that conventional radar technologies do not work well at low altitudes and are designed to detect “large chunks of metal” to put it crudely. To detect drones using radar technology, the radars must be placed at ground level and must work on millimetre wavelengths resulting in a short overall range. Aviation stakeholders have demonstrated to us that no one solution is effective in all cases and the general view among CNI assets has been that there can be no true “national picture” of a potential threat from drones using any one technology.
Therefore, the only effective solution, given that not all vulnerable assets are government controlled, is to build a real-time intelligence network where intelligence among many different types of short-range devices along with less accurate long-range devices are used to build an overall intelligence picture to be used to evidence for or against risk. The government ought to develop a voluntary scheme among the private sector and a mandatory scheme among CNI assets which is funded by the government.
The general consensus among CNI assets has been that electronic conspicuity is likely to be the future. Coupled with the government requirement that all drones are registered from the end of 2019, this would mean that drones would have to be fitted with devices which squawk a call sign associated with the drone registration. Manufacturers are also now adding the ability of drones to squawk basic instrument measurements from the drone and basic identifying information as in the case of transponders on conventional aircraft, meaning ATC can identify and track the flight path of a drone according to a flight plan filed if the drone needs to travel in controlled airspace or within the exclusion zone of key CNI assets such as airports. In this way, a malicious drone can be quickly identified if it appears to be unregistered, unchartered or veering off-course. Similarly, if manufacturers develop a standard for squawking basic flight information and drone identifiers, it will become very difficult to imitate a drone if for example it has a firmware identifier which is generated according to a robust algorithm as this cannot be copied.
PPL refers to a Private Pilot’s License issued by the UK Civil Aviation Authority to a non-professional pilot.
Whilst these real-time information systems are suggested, CNI assets were very keen to state the importance of the protection of freedoms for lawful drone pilots. They are less concerned about any area which is not controlled airspace because PPL pilots are able to fly freely in these areas without having to contact ATC in any event and so the relative safety risk posed is reasonably low.
Three of the key challenges we identified in conjunction with key critical infrastructure assets and aviation professionals were firstly that airports are presented with a dilemma when they are faced with a drone infringing controlled airspace. If they close, they are required to make a case that the airspace is clear and safe before they are legally allowed to re-open the airport. The position of aerodrome CNI assets in particular therefore, has become that airports should close at their peril.
Secondly, in aviation where there is a higher than usual threat from drones, we understand that pilots have no manual or guidance to follow if they encounter a drone mid-flight. Similarly in the case of aerodrome CNI assets, there is no joint consensus as to 5 W’s when deciding whether a particular drone does pose a threat. Third, if a drone does pose a threat, there is no legislation presently which enables a drone to be lawfully neutralised. Presently in aviation, risks to airspace are adjudged by multidisciplinary teams which are formulated on the initiative of key stakeholders, there is no national strategy on this and no specially ring-fenced funding for conducting this type of defensive exercise. It is important to note that given aerodromes are CNI assets, their decision as to whether a drone poses a risk to national security is strongly intelligence-led by the security services, in particular CCNI and therefore as the number of drones grows this is likely to outscale the resources of CCNI relatively quickly.
According to aviation professionals we interviewed, the current legislation in the UK is not adequate to ensure complete safety against attacks with malicious intent. There is in fact, no legislation entitling CNI assets or law enforcement to destroy malicious aircraft, the decision to destroy any aircraft infringing UK airspace is presently one for the military to take with the present status quo intended to address aircraft which is no longer controlled or dangerous aircraft with malicious intent such as a fighter jet conducting exercises in UK airspace as has been conducted by the Russian military in the past [14]. Whilst Police can and have shot drones out of the sky, and will not be held liable for doing so for public policy reasons, there is no legislation enabling this.
Our view, based upon consulting with CNI assets and aviation professionals, is that this is a decision that should be intelligence-led, depending on the circumstances of any threat. There are certain threats that no civil authority is equipped to deal with and so a military response would almost certainly be required.
However, in the case of the most common infringements and less sophisticated threats, our view is that this should be left to civil authorities which are properly equipped, funded and informed as part of a clear multidisciplinary process.
Anti-jamming solutions added to foreign governments (e.g., Russia, USA, Israel) research investment agendas in GPS-free and jamming-resistant drone technologies indicate that the UK Government should look beyond jamming devices for defending against attack drones. Our view is that the UK needs to design and build systems that can identify position and gather flight data of drones to intercept malicious attacks and allow enough time to destroy armed drones over safe places or even allow evacuation. Both the ‘capturing’ and jamming solutions rely on drone sighting for their operation. There is currently no fully integrated solution that can detect, identify, track and neutralise a drone. Hence, we also advocate the immediate deployment of logging devices in sensitive premises such as airports to collect flight data and record controller signals to help investigations in any successful breaches or infringements.
In April 2019, the UK Government launched a £2 million competition to counter the future threat of drones. Whilst this is a welcome step to develop the UK’s defensive capabilities for defeating hostile drones, we believe that this investment is not proportionate to the seriousness of the threat. Compared to the US Army investment of a $59 million program to develop GPS-independent drones alone or Russia’s $9 Billion investment in unmanned air vehicles by 2020, the UK Government is not investing enough to counter the increasing threat of armed drones by comparison.
Radio spectra are used by various aircraft equipment, particularly for communication.
From a technical perspective, the low technical readiness of current commercial anti-drone systems creates a multitude of problems depending on the environment they are deployed in. For instance, airports or military facilities already have many sources of high radio spectrum congestion which makes drone detection impractical if not impossible. Additionally, there is a high risk that signal detection/jamming solutions may adversely interfere with existing systems and communication infrastructure. Hence, anti-drone technologies must work with existing systems, processes and technologies without requiring major adjustment of existing infrastructure. The number, coverage range, type of sensors, communication technology, etc. depends on the facilities being protected. Accordingly, the viability of implementing permanent drone defence systems depends on the characteristics and requirements of the specific application. As these systems require a high level of human intervention to operate the equipment and respond to attacks, the overall cost of such systems becomes prohibitive. This is particularly important as drone technology is rapidly changing, rendering defence systems obsolete in a short period of time.
One of the biggest misconceptions that we have encountered is that drones are dangerous. Drones are not of themselves dangerous, they are dangerous when used as such. There are many lawful uses of drones which are a source of good and which are beneficial to society. We have seen Operation Zenith [15] in particular at Manchester Airport which was a well-publicised exercise whereby blue light drones were used during normal airport conditions without any interruption to the airport’s operations. This is good quality evidence that drones can operate as part of ordinary air traffic without causing nuisance or interruption, providing the right air traffic control services are available and there is protection against potential malice.
It is apparent that there is an acute lack of investment, particularly by HM Government, in developing high-quality solutions to target a wide threat matrix. CNI stakeholders have clearly indicated that there are a lack of proven solutions and there is a media hype around the potential danger of drones. It is clear that drones can be dangerous in a variety of circumstances, but the aviation sector does not believe in general that a drone is more dangerous than a bird. Equally, there is a relatively low risk of a sophisticated attacks by terrorists using drones because of the state of drone technology, therefore only capable actors such as nation states could use drones to leverage biological, chemical and nuclear agents to attack the United Kingdom.
Our concluding view is that the future of drone use is likely to require a real-time shared intelligence network using the technology that is available to provide evidence for or against a threat in order to decide how to act. Where action is required, Parliament needs to create legislation to enable civil authorities to act where appropriate to deal with potential threats originating from drones.
9 May 2019
[1] Goldman Sachs, “Drones: Reporting for Work,” 2019. [Online]. Available: https://www.goldmansachs.com/insights/technology-driving-innovation/drones/. [Accessed: 09-May-2019].
[2] Department for Transport, Taking Flight: The Future of Drones in the UK Government Response. 2019.
[3] Openworks Engineering Ltd, “SKYWALL100 | openworksengineering.com,” 2019. [Online]. Available: https://openworksengineering.com/skywall-100/. [Accessed: 09-May-2019].
[4] Battelle, “DroneDefender,” 2019. [Online]. Available: https://www.battelle.org/campaign/dronedefender. [Accessed: 09-May-2019].
[5] The Guardian, “Heathrow and Gatwick invest millions in anti-drone technology | World news | The Guardian,” 2019. [Online]. Available: https://www.theguardian.com/world/2019/jan/03/heathrow-and-gatwick-millions-anti-drone-technology. [Accessed: 09-May-2019].
[6] Intelligence Online, “RUSSIA : Moscow’s DARPA develops GPS-free drones,” 2019. [Online]. Available: https://www.intelligenceonline.com/government-intelligence/2018/10/24/moscow-s-darpa-develops-gps-free-drones,108329300-bre. [Accessed: 09-May-2019].
[7] Intelligence Online, “UNITED STATES : Pentagon seeks GPS-independent drones,” 2019. [Online]. Available: https://www.intelligenceonline.com/government-intelligence/2018/09/12/pentagon-seeks-gps-independent-drones,108323286-bre. [Accessed: 09-May-2019].
[8] K. C. Teh, C. C. Teng, A. C. Kot, and K. H. Li, “Jammer suppression in spread spectrum,” in Proceedings of IEEE Singapore International Conference on Networks and International Conference on Information Engineering ’95, pp. 220–224.
[9] Loh-Ming Li and L. Milstein, “Rejection of Narrow-Band Interference in PN Spread-Spectrum Systems Using Transversal Filters,” IEEE Trans. Commun., vol. 30, no. 5, pp. 925–928, May 1982.
[10] P. T. Capozza, B. J. Holland, T. M. Hopkinson, and R. L. Landrau, “A single-chip narrow-band frequency-domain excisor for a Global Positioning System (GPS) receiver,” IEEE J. Solid-State Circuits, vol. 35, no. 3, pp. 401–411, Mar. 2000.
[11] D. Wang, J. Li, W. Gong, and S. Wu, “Attitude aided space-time multi-beamformer anti-jamming approach for satellite navigation receiver,” in 2014 12th International Conference on Signal Processing (ICSP), 2014, pp. 368–372.
[12] M. Hannon, Shaung Feng, Hyuck Kwon, and Khanh Pham, “Jamming statistics-dependent frequency hopping,” in MILCOM 2016 - 2016 IEEE Military Communications Conference, 2016, pp. 138–143.
[13] J.-H. Kang and K.-J. Park, “Spatial retreat of net-drones under communication failure,” in 2016 Eighth International Conference on Ubiquitous and Future Networks (ICUFN), 2016, pp. 89–91.
[14] BBC News, “RAF jets scrambled to Russian military aircraft off Scotland - BBC News,” 2019. [Online]. Available: https://www.bbc.co.uk/news/uk-scotland-highlands-islands-47748034. [Accessed: 09-May-2019].
[15] Altitude Angel and NATS Services Limited, “Operation Zenith - Powered By GUARDIAN UTM - November 21st 2018,” 2019. [Online]. Available: https://www.operationzenith.com/. [Accessed: 09-May-2019].