Daunting Challenge of Drone Defense

INTRODUCTION

The proliferation of an unmanned aircraft/aerial system (UAS) spreads beyond military and remote control (R/C) hobbyist markets in the last decade, spearheaded by the Chinese company DJI beginning with the release of the DJI Phantom 1 in early 2013 [1]. As the capability of these aircraft has increased to meet the needs of new users, these platforms have been repurposed in unintended ways. Nefarious use of drones raises concerns for several key reasons—drones’ ability to circumvent the billions of dollars’ worth of physical security barriers installed around the world; their ability to carry malicious payloads, spy devices, or just innate mass; and their ability to allow the operator to deliver this payload remotely and pseudo-anonymously. Even more concerning is that this capability can be obtained instantly with commercial-off-the-shelf technology under $1,000 (e.g., a DJI Mavic Pro rigged with an improvised bomb in Turkey [2]). These systems have been designed to eliminate the learning curve, enabling nearly anybody to fly them with little-to-no previous experience. This article will discuss the means by which security entities are attempting to protect assets against malicious drones.

NOTE: For the purposes of this article, the terms drones and unmanned aircraft are considered interchangeable. The focus of this article is on small unmanned aircraft, regardless of the level of autonomy involved, that are becoming ubiquitous in recreational, commercial, and military zones. These aircraft can span consumer products, home-built R/C aircraft, and small military systems, such as those developed by AeroVironment (United States) and Aeronautics (Israel). The term UAS will be used to encompass not only the drone but additional hardware, including the ground control station, software, payloads, and supporting equipment.

CONCERNS

A major domestic concern is the intentional or careless interference with general and commercial aviation. For instance, the shutdown that occurred at Gatwick Airport in December 2018, where a small UAS flying over airport property suspended all flight operations for 33 hours, resulted in an economic cost of $64.5M [3]. The drone disturbance caused the airport to be steeped in chaos, confusion, and helplessness but was a relatively small threat to life and property; one could argue the stress caused by interrupting travel of thousands of people over the holidays was far more devastating than the physical damage the drone might have done except in an absolute worst-case scenario.

Other drone concerns include potential terroristic acts from protesters. In 2015, a protestor against the Japanese government’s nuclear energy policy flew a drone carrying radioactive sand onto the roof of the Japanese prime minister’s office. The drone was not discovered until 13 days after it was initially flown onto the roof. The radiation levels of the cesium source were low enough to not be harmful; however, the clandestine nature of the operation is disconcerting [4].

Although there are no successful cases of domestic terrorism with drones, one can wonder if a case such as the Austin, TX, serial bombings in 2018 would have led to more destruction if the suspect used drones rather than trip wires and FedEx to attack his targets [5]. The destructive potential for terrorist or militia use of small, commercially-available drones has been demonstrated throughout the world, from Syria [6] to Ukraine [7] and Venezuela [8] to the Philippines [9], and has been well documented in other press reports on the topic [10].

In addition to the threat of modified consumer drones for nefarious purposes, smaller UAS’s developed by defense industries are also rapidly increasing in capability and proliferation at levels of both peer competitors (the People’s Liberation Army reportedly recently acquired the CH-901 armed UAS [11]) and rogue regimes and their proxies (such as attacks on Saudi and Emirati civilian facilities by Iran and Ansar Allah [12]). For the foreseeable future, security forces across the globe should plan, in advance, on how to respond to a UAS threat, whether they are protecting assets on a battlefield or a civilian center.

FRIEND OR FOE?

One of the most significant issues with defense against a UAS is determining intent. This is, in part, due to the anonymity of the operator. Security forces responding to a situation like that at Gatwick Airport have little awareness whether the perpetrators are kids pulling a prank, a hobbyist hoping to upload exciting close-up aerial videos of commercial airliners during takeoff and landing, or someone nefarious, like the Austin serial bomber or an internet extremist trying to wreak havoc with a live audience [13]. Security forces must weigh the risks of overreaction vs. underreaction to a drone incident, as both can lead to loss of life or property.

Determining the intent of a UAS is challenging to do proactively. In the United States, the Federal Aviation Administration’s (FAA’s) response to the issue of ambiguity is focused around mandating remote identification and development of unmanned traffic management standards. This would help address the reckless and careless operators using UAS’s as long as they have properly configured the UAS to those security standards, a “papers, please” approach to interrogating drones and ground stations flying in public zones. (Note: there has been significant pushback against the FAA’s planned implementation of remote identification from industry and hobbyist groups, but this article will not address those issues.)

These efforts would allow security forces to detect, monitor, and, if needed, interdict some UAS’s interfering with public safety. More importantly, security forces will be quicker to respond to UAS’s flying without proper remote identification certifications [14]. With the careless and reckless operators thus handled, all other UAS’s can be treated as nefarious, akin to driving a car with no license plate, and therefore cue security forces to react more quickly with additional means of response.

Security operators are typically left with analyzing the behavior of a drone to determine its intent, such as evaluating its track history and current trajectory. Tracking solutions typically involve radar systems that require continuous operation and constant monitoring. The operator would have to interrogate whether a given track was indeed a UAS and then analyze the target trajectory to make an informed decision as to whether it was hostile. This capability does provide situational awareness of threats in the area, even if the threat UAS is not interdicted.

FALSE ALARM FATIGUE

Even with all means of response authorized, defense against drones is far from an easy task. Security forces responsible for detecting and responding to malicious drones suffer the same issues of fatigue and vigilance and boredom and paranoia as all sentries. A security officer could go weeks, months, years, or even a career without encountering a malicious drone; meanwhile, the security officer may be bombarded with false or nuisance alarms in that same time frame. Security organizations are left choosing between highly-sensitive (and expensive) drone defense systems that can cause hundreds of false alarms per week vs. less-sensitive systems that may miss the very threats they are supposed to detect.

The most significant impact that can be made to current systems is including operator-assisting algorithms to aid the operator in interpreting the system data presented to them. Significant investment into automated sensor processing is necessary to accurately parse the data to minimize the false alarms or nuisance alarms presented to the operator. Further investment into human systems integration can focus on minimizing operator workload by efficiently presenting the data to the operator. For example, radar displays can be overwhelmed by nuisance alarms, especially when small UAS’s can have similar radar cross-section values as large birds, to an extent that obfuscates the presence of a UAS on the radar display. While difficult for an operator to sift through in the required time, automated intelligent processing to neglect these nuisance alarms would alleviate this issue. All other sensors have similar nuisance alarms or background noise issues.

Optic sensors can have nuisance alarms from birds and background noise from clouds or terrain, while electronic detection sensors can be cluttered by extraneous signals or must overcome a high ambient noise floor, and acoustic systems can easily be saturated by background noise. Advancements to shift away from operator dependence and put the burden on automated processing to detect low signal-to-noise ratio signals and reduce the number of notifications from nuisance alarms will simplify the problem to something the operator can easily manage.

NO SILVER BULLET

Once detected and assessed to be a threat, security forces must then employ a defeat mechanism similar to the one shown in Figure 1. The most prevalent is electronic interference/attack (barrage jamming) of the UAS command and control signals to disrupt the operator control and initiate a fail-safe mode. The emission of radio frequency (RF) energy by the counter (C)-UAS system raises the RF noise floor in the surrounding area, causing the UAS command signals to be lost in the noise. This results in the UAS entering a hover/loiter, landing, returning to its takeoff location, or continuing its flight path with a significant capability handicap, much like when a UAS flies too far away from the ground station transmitter. In some cases, the Global Positioning System (GPS) link may be jammed and cause the aircraft to enter an attitude hold in which it maintains current trajectory, attempts to use less-reliable navigation sensors, or hovers in place, depending on the selected fail-safe.

A more precise mitigation technique can be used to avoid electronic fratricide, which is narrowband jamming. This technique relies on jamming the precise frequencies on which the command and control signal will hop. To achieve this, the C-UAS system needs to have a threat library, which includes information on these hopping signals.

There are issues relying on a threat library for UAS detection and interdiction. Constant reverse engineering on datalinks is required to maintain a current library, which involves cost and effort. The evolution of the DJI datalink is a paradigm of why threat libraries are difficult to maintain. The early DJI Phantom models relied on Lightbridge technology, which was a hybrid between hardware and software for the transmission system. However, newer models utilize OcuSync, which is strictly software-defined radio (SDR) based [16]. OcuSync 2.0 firmware is upgradable through patches and can automatically switch between multiple bands for communication (2.4 GHz or 5.8 GHz industrial, scientific, and medical bands). The agility at which newer SDR-based systems can alter their frequency band and hopping pattern will cause difficulty in keeping a consistent library. The constant evolution and development to make more robust datalinks for UAS by the commercial industry will require constant efforts to update the threat libraries by the counter-drone industry.

Barrage jamming and narrowband C-UAS systems can be a promising mitigation tool against UAS flying by command and control links. But what happens when RF links are not present? UAS researchers are making flight operations more autonomous, less reliant on active datalinks, and more reliant on GPS-based or vision-based navigation. Additional efforts have explored the utilizing long-term evolution networks as the backbone for the communications link [17]. In both cases, jamming these frequencies will have either no effect or effects with significant, unintended consequences of electronic fratricide (i.e., blocking cellular, Wi-Fi, and GPS service) to the nearby population. The FAA has testified to Congress their hesitance for many security forces to be given the ability to jam the command and control and GPS signals of a drone, as the solution (jamming) is worse than the problem (presence of an unauthorized drone) in many scenarios [18].

HACKIN’ AIN’T EASY

A more effective and surgical means of counter-drone defense is to try to hijack the drone by hacking its control system, thereby allowing the security responder to land the drone in a safe area, with minimal risk of disrupting bystanders. However, this methodology has legal issues. The Fourth Amendment of the U.S. Constitution; U.S. Code: Title 18 Electronic Communications Interception; and U.S. Code: Title 50 Collections, Stipulations, and Limitations of the Preventing Emerging Threats Act of 2018 suggest this methodology in U.S. jurisdictions is applicable only after other methodologies fail [19]. There are also technical hurdles to overcome. Just as the efficacy of a flu vaccine is dependent on the strain of flu, a cyber-based countermeasure system is highly dependent on the strain (software and firmware version) of the drone and whether counter-drone engineers have amply developed an effective attack for that drone. Rapid improvements toward more resilient command and control protocols made by drone manufacturers compound this problem.

DJI, the leading UAS manufacturer that dominates ~70% of the commercial UAS market [20], has developed its own cyber-based drone defense system, Aeroscope. This system exploits back doors in the communications and flight control system to allow security officials to hijack nearby DJI drones. Future collaboration between security and regulatory officials and drone manufacturers can provide more thorough portfolios of hackable drones, although it is unlikely counter-drone systems will be loaded with hacks to attack every potential threat, and particularly drone-savvy individuals are able to build and program custom UAS that have no communication systems to hack at all.

SHOOT TO KILL?

When electronic interference or hacking the command protocol fail to disrupt an unauthorized drone, security forces then require a “hard kill,” such as nets, bullets, missiles (including drone-to-drone intercepts akin to the one shown in Figure 2), lasers, and high-power microwave bursts (shown in Figure 3). Hard-kill systems, if they hit the target, are less discriminating than the barrage-jamming and cyber techniques mentioned previously. However, they come with other issues, such as collateral damage, environmental interference, and range to target, that can degrade or negate their suitability. Hard-kill defenses have their own safety risks that must be mitigated and do not guarantee an optimistic outcome. Those safety risks—hitting something other than the target or bringing down a drone in an uncontrolled, destructive crash—have restricted fielding of hard-kill systems primarily to conflict zones [21].

YOU GET WHAT YOU PAY FOR

The best approach is to use a combination of sensors—radars, cameras, and signals intelligence devices—to increase the odds of detecting and defeating a UAS. Using a system-of-systems approach can leverage capabilities inherent in each modality while mitigating weaknesses. Sensors can correlate information to weed out false alarms and maximize probability of an effective defense. Similarly, a layered defense approach can use a combination of defeat options—jammers and interceptors—to improve probability of UAS interdiction. A perimeter defense solution can be employed with a network of sensors and efforts to maximize the coverage of the protected area. An example of a system of systems is as follows: an electronic support node can be used to detect a nearby threat UAS emitting a signal early. A radar system can work with an optic for slew to cue to achieve positive threat identification, then an electronic attack system can attempt to jam to the UAS datalinks. If this proves unsuccessful, then a kinetic solution can be used to interdict the aircraft. The overlap of sensors ensures that the limitations of any given sensor are mitigated by the capabilities of the others.

The difficulties with a system of systems approach are two-fold. First, the monetary cost of fielding radars and cameras and net guns and jammers will strain budgets. Second, security officers armed with an arsenal of sensors and effectors may be overwhelmed with data and engagement decisions. However, decisions related to selecting engagement options, threat assessment, and potential collateral damage must be made in less than a minute.

Being able to successfully detect and defeat a worst-case drone threat will necessitate that much of the decision making in the engagement be automated via the C-UAS system. This ultimately means risks of overreaction or underreaction to drone incidents are left to system developers and parameters set by each security unit rather than a security officer.

WHAT’S THE SOLUTION?

Each scenario or operation will have a different optimal C-UAS solution. As such, numerous C-UAS systems have been developed based on specific missions and operational needs from DoD commanders over the last several years. Depending on the mission or protected area, the appropriate C-UAS system will have an optimal size, weight, and power (SWaP). Systems can be segregated into dismounted, mobile, and fixed-site configurations.

Dismounted systems might include electronic detection and attack capabilities for point defense but must be light enough to be carried in a backpack for long periods. Electronic detection and attack systems offer the only low SWaP solutions feasible. Optic systems could meet SWaP considerations but have not yet been proven useful in dismounted operations; identification and targeting issues would only be compounded by a moving operator. Similarly, acoustic systems (consisting of a few microphones) could be worn by an operator. Although some dismounted solutions for acoustic-based gunshot detection have been developed, there has not been adequate exploration of this capability for a C-UAS to determine operational feasibility.

Mobile systems, such as the one shown in Figure 4, must be small enough to mount onto a vehicle, rugged enough to sustain shock and vibration loads, and able to operate off of generator power. However, there is more flexibility in the types of systems that can be applied toward the problem compared to dismounted systems. Electronic detection systems can be present and used for early warning systems. Electronic attack systems can radiate more power because they can operate off generator power instead of batteries. Smaller radar systems can be utilized on mobile vehicles [24]. Depending on the rules of engagement, visual identification may be required before engagement. As a result, optics should be integrated with any radar systems on a mobile solution. Acoustic systems still have not made much headway when it comes to mobile operations, as the noise floor caused by generator noise or vehicle movement will mask the target signal.

Lastly, kinetic defeat solutions can be present on mobile platforms. Interceptors have become more common, both in the form of multirotor UAS and tube-launched systems, such as AeroVironment’s Switchblade and Raytheon’s Coyote, which use onboard sensors for guidance to the threat UAS. Laser systems have also made recent advancements to find their way into the battlespace. There are deconfliction issues using high-energy lasers (HELs) in the battlespace; however, systems like MEHEL and CLaWS are overcoming these and provide a mobile HEL solution [26, 27].

All the same sensors and effectors applicable to mobile operations can also be employed for fixed sites, with the added benefit of extra space, infrastructure for installation, and use of shore power, thus eliminating many SWaP concerns. Detection nodes can be dispersed around the perimeter of the defended area and placed so the electronic or acoustic detection nodes are isolated from interference. Effectors can be dispersed or centralized, depending on the defended area, and lines of fire can be cleared during the planning process. Command and control systems are critical because networked sensors and effectors need to communicate with each other in a timely manner to enable the kill chain and provide a common operational picture.

In December 2019, the Undersecretary of Defense for Acquisition and Sustainment designated a Joint C-UAS Office and selected the Army as the lead Executive Agent in charge of assessing currently-fielded C-UAS programs for the DoD [28]. Assessments of several C-UAS systems are ongoing to determine the best-of-breed systems for use in various operational conditions. The Joint Office aims to leverage the efforts and expertise of each Service to improve current C-UAS capabilities. However, while the DoD is becoming more organized to develop, test, and field C-UAS systems, the rest of the federal government and state and local entities will have a far more piecemeal approach. One can easily imagine police departments in neighboring jurisdictions having completely different C-UAS equipment, training, and policies.

CONCLUSIONS

Protecting assets or personnel from nefarious UAS is an ever-increasing problem due to capability improvements and unhindered proliferation. Whether the drone operator is a terrorist, foreign intelligence agent, or a kid innocently flying a toy, security forces must be prepared to identify and proportionally respond to the threat. There is no silver bullet for the drone defense problem set, so continued investment, testing, and improvements to counter-drone technologies are necessary. The complexities and trade-offs of this problem must be carefully navigated to effectively manage policy, technical challenges, and funding restrictions to ensure adequate defense capabilities.

REFERENCES

1. DrDrone.ca. “Timeline of DJI Drones: From the Phantom 1 to the Mavic Air.” https:// www.drdrone.ca/ blogs/drone-news-drone-help-blog/timeline-of-dji-drones, accessed 1 April 2020.
2. Acar, C. “3 Types of Pkk ‘Armed’ Drones Downed by Security Forces in Turkey’s SE Areas.” https://twitter.com/Acemal71/ status/934788399441088512, accessed 1 April 2020.
3. Calder, S. “Gatwick Drone Disruption Over Christmas Cost £50m.” https://www. independent.co.uk/travel/news-and-advice/gatwick-drone-airport-cost-easyjet-runway-security-passenger-cancellation-a8739841.html, accessed 1 April 2020.
4. Murai, S. “Man Who Landed Drone on Roof of Japa­nese Prime Minister’s Office Gets Suspended Sentence.” https://www.japantimes.co.jp/news/2016/02/16/na­tional/crime-legal/man-landed-drone-roof-japanese-prime-ministers-office-gets-suspended-sentence/#.XsLVu3d­FymQ, accessed 1 April 2020.
5. Weber, P. J., and W. J. Weissert. “Fear Mounting in Austin as Serial Bomber Uses Tripwire Along a Street.” https://www. denverpost.com/2018/03/19/austin-bomb­ing-spree/, accessed 1 April 2020.
6. PressTV. “Russian Military Eliminates Militants Be­hind Syria Air Base Attack: Video.” http://french.presstv.com/Detail/2018/ 01/12/548707/Russian-military-elim­inates-militants-behind-Syria-air-base-attack, accessed 1 April 2020.
7. Mizokami, K. “Another Ukrainian Ammo Dump Goes Up in Massive Explosion.” https://www.popularmechanics.com/ military/weapons/news/a28412/ukrainian-ammo-dump-explosion/, accessed 1 April 2020.
8. Koettl, C., and B. Marcolini. “A Closer Look at the Drone Attack on Maduro in Venezuela.” https://www.nytimes.com/ 2018/08/10/world/americas/venezuela-video-analysis.html, accessed 1 April 2020.
9. Tima, R. “Marawi, The Drone War.” https://www. gmanetwork.com/news/news/ specialreports/632793/marawi-the-drone-war/story/, accessed 1 April 2020.
10. Rassler, D. “The Islamic State and Drones: Supply Scale, and Future Threats.” https://ctc.usma.edu/wp-content/uploads/ 2018/07/Islamic-State-and-Drones- Release-Version.pdf, accessed 1 April 2020.
11. DefenseWorld.net. “China to Buy 2 Types of Loitering Munitions.” https://www. defenseworld.net/news/26507/ China_to_Buy_2_Types_of_Loitering_Muni­tions#.XsLbTndFymQ, accessed 1 April 2020.
12. Hubbard, B., P. Karasz, and S. Reed. “Two Major Saudi Oil Installations Hit by Drone Strike, and U.S. Blames Iran.” https:// www.nytimes.com/2019/09/14/ world/middleeast/ saudi-arabia-refineries-drone-attack. html, accessed 1 April 2020.
13. Boot, M. “Why Social Media and Terrorism Make a Perfect Fit.” https://www. washingtonpost.com/opin­ions/2019/03/ 16/why-social-media-terrorism-make-perfect-fit/, accessed 1 April 2020.
14. Stubblefield, A. “Drone Security: Enhancing Innova­tion and Mitigating Supply Chain Risks.” https://www.transportation. gov/testimony/drone-security-enhancing-innovation-and-mitigating-supply-chain-risks, accessed 1 April 2020.
15. Carter, J. “Joint Base MDL Implements New Drone­buster Technology to Deter Evolving.” https://www.jbmdl.jb.mil/ News/Article-Display/Article/2076897/joint-base-mdl-implements-new-dronebuster-technology-to-deter-evolving-threat/, accessed 27 March 2020.
16. djibestdrones.com. “dJI OcuSync 2.0: What You Need to Know About This FPV Transmission System.” http://djibestdrones. com/dji-ocusync-2-0/, accessed 1 April 2020.
17. Kannan, K. “Nokia Showcases LTE Technology for the Use of Drones in Smart Cities.” https://www.nokia.com/about-us/ news/releases/2016/02/04/nokia-show­cases-lte-technology-for-the-use-of-drones-in-smart-cities/, accessed 1 April 2020.
18. National Cable Satellite Corporation. “Federal Avia­tion Administration Oversight Hearing.” https://www.c-span.org/video/?463129-1/federal-aviation-administra­tion-oversight-hearing, accessed 1 April 2020.
19. Rupprecht Law P.A. “7 Big Problems with Counter Drone Technology.” https://jrupprechtlaw.com/drone-jammer-gun-defender-legal-problems/, accessed 1 April 2020.
20. Business Insider Intelligence. “Commercial Un­manned Aerial Vehicle (UAV) Market Analysis – Industry Trends, Forecasts and Companies.” https://www.busines­sinsider.com/commercial-uav-market-analysis, accessed 1 April 2020.
21. Pappalardo, J. “The Air Force Is Deploying Its First Drone-Killing Microwave Weapon.” https://www.popular­mechanics.com/ military/weapons/a29198555/phaser-weapon-air-force/, accessed 1 April 2020.
22. Vergun, D. “Proliferation of Drones Posing Risk for U.S. Military, Expert Says.” https://www.433aw.afrc. af.mil/News/Article-Display/Article/1612325/prolifer­ation-of-drones-posing-risk-for-us-military-expert-says/, accessed 27 March 2020.
23. Raytheon Technologies. “Phaser High-Power Micro­wave System.” https://www. raytheonmissilesanddefense.com/ capabilities/products/phaser-high-power-microwave, accessed 25 March 2020.
24. Mizokami, K. “This Is the ATV-Mounted Jammer That Took Down an Iranian Drone.” https://www.popularme­chanics.com/ military/weapons/a28471436/lmadis-iranian-drone/, accessed 1 April 2020.
25. Swanbeck, D. “Program Executive Officer Land Sys­tems.” https://www.marcorsyscom. marines.mil/PEOs/- PEO-LS/PM-GBAD/, accessed 27 March 2020.
26. U.S. Army Space and Missile Defense Command. “Mobile Experimental High Energy Laser (MEHEL).” https://www. smdc.army.mil/Portals/38/Documents/ Publications/ Fact_Sheets/MEHEL.pdf, accessed 1 April 2020.
27. Mizokami, K. “U.S. Marines Have Started Testing Its Drone-Killing Laser Weapon.” https://www.popular­mechanics.com/ military/research/a28120435/laser-weapon-claws/, accessed 1 April 2020.
28. Tadjdeh, Y. “Defense Department to Stand Up New Counter-Drone Office.” https://www.nationaldefen­semagazine.org/ articles/2020/1/14/just-in-defense-department-to-stand-up-counter-drone-office, accessed 1 April 2020.