Homeland security planners face a daunting array of conceivable threats that terrorists might wish to perpetrate against a multitude of targets. If there is any lesson that 9/11 provides it is that security planners must avoid confusing unfamiliar threats with improbable ones. In the immediate aftermath of 9/11, security planners responsible for protecting the Pentagon admitted that they worried about familiar airborne threats rather than hijacking large commercial airliners employed as suicide weapons. For example, the October 1994 Cessna that crashed into the White House figured heavily in their scenario planning for an attack on the Pentagon. A scenario involving the hijacking of commercial airliners used as human-guided cruise missiles looked strange and thus improbable. Planners were simply much more familiar with airplane hijackings being used to negotiate the release of individuals rather than employed as terror weapons.
Likely threats appeared to clash with more familiar ones on January 10, 2002, when, just four months after 9/11, the CIA released the "National Intelligence Estimate" on the ballistic missile threat to the United States. In it, the intelligence community argued that the United States is more likely to be attacked with weapons of mass destruction (WMD) using "ships, trucks, airplanes or other means" than with an intercontinental range ballistic missile. The 2002 National Intelligence Estimate (NIE) also stated that cruise missiles offer a better alternative than ballistic missiles if launched from forward areas, with the most plausible forward deployment scheme being the use of a covertly equipped commercial vessel. Although the intelligence community had state actors principally in mind in reaching this latter judgment, terrorists and non-state actors have also figured into American planning. In a 2001 counter-terrorism exercise, the North American Aerospace Defense command (NORAD) simulated a cruise missile launch from a merchant ship in the Gulf of Mexico, causing the NORAD test director to observe, "we are naked ... [and] have no capacity to deal with that kind of problem."
The 34-nation Missile Technology Control Regime (MTCR)—the only extant multilateral supply-side arrangement covering the transfer of missiles, related equipment, material, and technology relevant to WMD delivery—found reason to worry about terrorist access to UAVs shortly after 9/11. At its 2002 annual plenary meeting in Warsaw, MTCR member states took the first step toward addressing possible terrorist use of UAVs and cruise missiles by committing to examining ways of limiting the risk that controlled items and their technologies could fall into the hands of terrorist groups and individuals. Two and a half years after this commitment, it seems appropriate in this Issue Brief to examine how real or imagined the threat of terrorist use of UAVs (including cruise missiles) might be, and to the extent that such a threat appears real, to pose the question of what steps might be taken to lessen the threat's emergence.
Two particular scenarios of relevance to possible terrorist use of UAVs are worthy of examination here. Besides differing in their complexity of development and execution, they are also unlike each other in their launch point of origin. The first entails launching a converted anti-ship cruise missile—transformed into one capable of flying over land—from a covertly equipped commercial vessel against a U.S. city or other target. Thus, the missile threat would emanate from outside of U.S. borders and face whatever aircraft NORAD could muster in time, assuming that such a low-flying missile were detected in the first place. The $9 billion a year that the United States spends on defenses against ballistic missiles does nothing to address this threat. The first scenario roughly equates to the one simulated in the 2001 NORAD counterterrorism exercise and stated in the 2002 NIE (albeit in the state-actor context). The second case focuses on converting a small aircraft into an armed UAV and launching it from a domestic point of origin against its intended target. Like al-Qai'da's exploitation of lax airport security on 9/11, this scenario would exploit extant shortcomings in U.S. capabilities to detect and intercept threats flying under the Federal Aviation Administration's (FAA) radar system (<3,000 feet). Indeed, terrorists seem to be adjusting to 9/11's positive effects on airport security. According to the director general of intelligence for Canada's armed forces, terrorist groups have already purchased ultra-light aircraft and hang-gliders to work around effective security precautions against hijacking large commercial airliners.
Case 1: Launching a Cruise Missile from a Covertly Equipped Commercial Vessel
The possibility of terrorist, as opposed to state actor, involvement in a cruise missile scenario is not purely hypothetical or based only on the NORAD counterterrorism exercise. According to two former National Security Council staff members, al-Qai'da is believed to possess 15 freighters. This has raised concern that terrorists could manage to acquire, deploy, and successfully launch a cruise missile from just outside U.S. waters using a freighter, for example, as the launch platform. But such a proposition, fortunately, faces a number of obstacles. Foremost is the supposed ease with which just about any person or small group could build a simple, self-guided cruise missile at minimal expense and based entirely on off-the-shelf component technologies. This proposition was given light during 2003 by a New Zealand engineer, Bruce Simpson, who created a website with the sobriquet "Do-It-Yourself Cruise Missile." His objective was to document (and publish) his progress in building such a simple cruise missile in his garage for under $5,000. Before having the chance to test his product, the New Zealand government, under pressure from the United States, forced Simpson to shut down his efforts. Simpson told BBC News that he had proved "that by using off-the-shelf technology in a suburban garage a terrorist can create a weapon against which there is no effective defense."
But before too much is made of do-it-yourself cruise missiles, it is important to note that it is by no means clear that Simpson's efforts would have proven successful. Just because individual component parts are available from commercial vendors does not mean that they can readily and successfully be integrated to produce a reliable system. Such a notion smacks of technological determinism or a simple reductionist viewpoint, whereby technology diffuses easily and smoothly into complex systems. A more plausible explanation for achieving success with any military system, no matter how simple it may appear on the surface, depends on certain tacit knowledge skills, in this case, system engineering or integration that is required to integrate actuators and servo mechanisms crucial for moving the cruise missile's control surfaces based on commands from a flight management computer. Simpson's technical approach to the flight management task belies the ease with which it can be accomplished reliably. It is also important to note that Simpson's small cruise missile represented a proof-of-concept design, which involved only a few kilograms of payload.
Rather than building a cruise missile from scratch, it would appear more sensible to convert one of many widely available anti-ship cruise missiles, a proliferation course that Iran is currently thought to be following and one that Iraq attempted prior to the second Gulf War in March, 2003. In Iran's case, it is reported to be upgrading around 300 Chinese HY-2 Seersucker anti-ship cruise missiles by fitting them with a turbojet engine and new guidance systems. Using a surplus Chinese Seersucker missile has certain distinct advantages, not least its wide availability. From a weapons-proliferation standpoint, the Seersucker/Silkworm anti-ship cruise missile shares the Scud ballistic missile's ubiquity: it is available globally, even in such countries as Bangladesh, the Democratic Republic of Congo, Dubai, Egypt, Iran, Iraq (until 2003), North Korea, and Pakistan. Given their age, many surplus missiles are likely available at modest prices.For states, turning anti-ship cruise missiles into ones that can attack targets on land is nothing new. The U.S. Navy has converted the Harpoon anti-ship cruise missile, which is exported to 24 nations, into the Stand-off Land-Attack Missile (SLAM). Russia has done the same with its export family of Klub anti-ship missiles, one of which is being jointly produced by India and Russia under the name Brahmos. Yet, converting these modern anti-ship cruise missiles, which are densely packed with electronics and comparatively much smaller than the Seersucker/Silkworm family of missiles, offers little space for adding fuel to extend the range much beyond 100 km. Once the original autopilot and radar are removed from the Seersucker, for example, there is significant space for additional fuel to propel the missile (assuming it uses a suitable turbojet engine) to around 1,000 km.
Thus, two challenges stand in the way of converting an anti-ship cruise missile into a land-attack system: finding and installing a suitable turbojet engine to replace the existing liquid rocket engine and replacing the anti-ship guidance and control system with a land-attack one. Achieving the first challenge permits the missile to travel beyond a mere 85-100 km, which would enable the missile to be launched from a ship lying significantly outside U.S. territorial waters. The second is necessitated by the different challenges of guiding a cruise missile over the flat sea compared with over the more variegated surface of the earth. The former seaborne setting means that the anti-ship cruise missile can use a simple autopilot for navigation and a terminal radar guidance system to seek out a large metal object (a ship). Were such a guidance system used over land, it would encounter difficulty hitting its intended target, which would normally furnish far less contrast than a ship at sea. And because cruise missiles intentionally fly low to avoid detection, they must negotiate over and around varied terrain (mountains and man-made objects, for example) before they reach their intended targets. These demands dictate the use of a more sophisticated guidance and navigation system.
By no means should the challenges of converting an HY-2 Seersucker into a land-attack cruise missile be seen as modest. Albeit limited in time, Iraq's experience with the process illustrates the demands. Called the Jinin project, Iraq conceived of a program to extend the range of the Seersucker missile to 1,000 km in November 2001. The program benefited in part from their previously successful efforts to extend the range of the Seersucker from about 100 km to 150 km. The Jinin project was approved in June 2002, but for fear of being discovered by UN inspectors upon their return to Iraq, the project was abandoned in December 2002. Yet, Iraqi engineers indicated that the program would require three-to-five years of development time, integrating the work of several research, development, and production organizations. During the project's short-lived initial phase, the development team used computer simulations to assure the structural integrity and stability of the missile due to the installation of a new propulsion system. Using turbine engines from surplus Russian Mi-8 helicopters, Iraqi engineers encountered problems achieving enough thrust and turned to another more advanced helicopter turbine engine. The senior program manager for the project said that a flight test could be conducted in three years. No work was accomplished on the guidance and control system, as the Iraqis initially believed that the existing HY-2 guidance system could be used in the Jinin project. They eventually did recognize that a foreign global positioning system (GPS) system would have to be acquired and integrated into a flight management system, but decided to await completion of flight tests proving that the Jinin's new propulsion system worked satisfactorily without destabilizing the vehicle's flight characteristics. This latter consideration was prompted by the realization that all of the internal rearrangements of the missile's subsystem components (perhaps even including additional fuel bladders) would likely affect flight control and stability.
Launching a modified Seersucker from a floating vessel presents yet another, if less difficult problem. The missile would be nearly 8 meters long with a 2.4 meters wide wingspan and a weight of nearly 3,000 kg. Nonetheless, a standard 12-meter commercial shipping container is more than adequate to house the missile. However, a small erector would have to be constructed to permit the missile to rise from the container for launching. Moreover, a solid rocket booster would be required to propel the missile from its erector to a height sufficient to commence flight via the turbojet engine.
Taken together, these tasks are beyond the capabilities of small terrorist groups, unless they could arrange various shortcuts amounting to virtually wholesale dependence on foreign assistance from a state missile program. Moreover, there are numerous critical failure points along the path to completing this complex project, daunting even for state programs such as Iraq's (pre-2003) and contemporary Iran's. If a terrorist group were set on using a UAV for weapon delivery, it is far more likely that significantly less sophisticated technical paths would be chosen.
Case 2: Converting a Small Airplane into a Weapons-Carrying UAV
Concern about the use of small UAVs as terrorist weapons began shortly after 9/11. By the following summer, Secretary of Defense Donald Rumsfeld reportedly sent the White House a classified memorandum warning about the spread of cruise missiles and UAVs and possible terrorist use of rudimentary technology to attack the homeland. By early 2003, senior Bush administration officials told Wall Street Journal reporters that the president was keenly interested in intelligence reports that Iraq was developing small, easily transportable UAVs as weapons that could be shipped into the United States or built here and then used to disseminate chemical or biological agents. Iraq admitted to the UN that it had sought to convert manned aircraft, such as the MiG-21, into a chemical-biological weapons (CBW) platform in the early 1990s, and there is indirect evidence that the L-29 trainer aircraft, which also was converted into a UAV, may have been intended for CBW delivery. But the Iraqi Survey Group found in 2003, after Operation Iraqi Freedom, that Iraq's small UAV (known as Al Musayara-20 and thought to be capable of disseminating biological agents) had not yet been equipped to do so. However, the Al Musayara-20 had the range (>500 km), payload (20 kg), guidance (GPS-aided), and autonomy needed to be an effective biological delivery means, if the Iraqi leadership had decided to pursue such a course. But it is one thing for a state actor like Iraq, depending as it did on extensive foreign assistance, to develop such small UAVs for weapons delivery, and quite another for a terrorist group to accomplish the same task. Yet, with enough outside assistance and access to commercially available subcomponents, it is at least conceivable, if not highly likely, that a terrorist organization could develop a much cruder version of Iraq's UAV exploits.
Acquiring an ultra-light or kit-built recreational airplane represents perhaps the most worrisome terrorist course of pursuing an unmanned attack weapon. The distinction between an ultra-light and larger kit-built airplanes is primarily one of size--according to FAA regulations, a powered ultra-light cannot exceed 254 pounds of empty weight, cannot carry more than 5 gallons of gasoline, nor exceed more than 55 knots at full power in level flight. At first blush, one might assume that any terrorist group connected to al-Qai'da would not require an unmanned vehicle. Given the jihadist's fascination with suicide, which combines an element of romanticism with sacrifice and exculpation, there are ample candidates to fly such an aircraft to its intended target. But attack effectiveness could supersede suicidal fascination. The reason is straightforward: if a human is not piloting the aircraft, attack effectiveness could be greatly enhanced, as the space otherwise devoted to a pilot and co-pilot can be given over to payload, conventional or otherwise. Even an ultra-light aircraft could carry 150 or more pounds of payload, more than enough to mount a suitable sprayer for a biological agent or a load of conventional explosives. There is little doubt that al-Qai'da is interested in acquiring nuclear, biological, or chemical weapons, and has discussed ways of procuring radiological materials for a "dirty" bomb. Radiological dispersal becomes possible and potentially effective with a UAV over large urban areas, but only if the source material is cesium chloride--the one radiological source that comes in powdered form. Dispersal of chemical or especially biological agent is ideally suited for a UAV; its flight stability permits the release of agent evenly along a line of contamination. As for conventional payloads, while a small UAV doesn't begin to compare with a jumbo jet's 60 tons of fuel, it is important to note that even gasoline, when mixed with air, releases 15 times as much energy as an equal weight of TNT. Thus, even very small planes carrying an extra large fuel tank in place of a pilot could do significant damage in an urban setting.
Beyond their potential effectiveness, another critical factor that is likely motivating terrorist interest in UAVs is the ease with which these systems can exploit the intended target's vulnerabilities. Surely, that is what compelled al-Qai'da to hijack large commercial airliners and turn them effectively into cruise missiles. The terrorist group, Hizballah, has twice (most recently, on April 11, 2005) flown an Iranian-furnished UAV over Israeli territory, surprising Israeli Defense Force radars with its ability to avoid detection and interception. The abysmal state of U.S. defenses against low-flying UAVs and cruise missiles invites terrorists to focus on acquiring UAVs as terror weapons. A July 8, 2004 House of Representatives hearing drew grim attention to the lax state of defenses against low-flying objects by examining the near-catastrophic circumstances surrounding the June 9, 2004 funeral for President Ronald Reagan. As officials gathered in the nation's capitol, a combination of human error, onboard technical malfunction, and computer incompatibility between the FAA and the Transportation Security Administration (TSA) caused security personnel to mistake the governor of Kentucky's official airplane as a terrorist threat. This led to the evacuation of hundreds of officials, the dispatch of two F-15 interceptors, and circumstances that nearly prompted the top general of the North American Aerospace Defense Command to order the governor's plane shot down. This near disaster in Washington, D.C., points up the inadequacy of the nation's capacity to identify friend from foe. Thousands of small private aircraft continue to wander into restricted airspace around Washington, D.C. annually. In the case of the governor's plane, this led to erring on the side of caution. But the fact that a disastrous mistake nearly ensued could produce an even more tragic result: inaction in the face of a genuine terrorist threat. In any event, the head of the U.S. Department of Homeland Security's directorate in charge of air defense has admitted that the current system may not be able to stop a determined adversary. An intelligent and committed terrorist is unlikely to fly a small airplane, whether manned or unmanned, above an altitude of 3,000 feet, where the FAA's radars would be able to detect and query the aircraft's transponder to establish its intentions. Clearly, terrorist use of a UAV would present severe warning, detection, and interception challenges.
Procuring a suitable aircraft is a vastly simpler task than turning it into an autonomous or even remotely controlled vehicle. By one accounting, the kit airplane market consists of nearly 100,000 copies of over 400 different designs from a worldwide list of manufacturers. An alternative course of action would be to use mini-UAVs, or model airplanes that can either be remotely controlled from the ground or programmed for autonomous flight. According to press accounts, al-Qai'da considered using remotely controlled airplanes packed with explosives to attach the G-8 Summit in Genoa, Italy in 2001. To be sure, working with smaller model aircraft reduces the chances of detection, but the limited amount of payload such models carry (a few pounds, at most) is a limitation. Kit airplanes, on the other hand, furnish between 150 and 300 pounds of payload and can be constructed in a single-car garage. Flying them autonomously (by programming route ways points into a flight management computer) rather than controlling them via line-of-sight radio commands from ground presents the greatest danger of detection due to the greater possibility of an accident. However, kit planes do not require an airport for launch and return; they can operate from flat fields in remote areas, thus reducing if not eliminating the chances of detection.
Unlike the conversion of a first-generation anti-ship cruise missile into a longer-range land-attack system, kit airplane conversion does not require a new propulsion system. The existing small reciprocating engine is more than sufficient to fly the craft to ranges of more than 500 miles. The stiffest challenge lies with integrating and properly installing and testing the various components needed to manage stable autonomous flight to a precise point on the ground.
Achieving successful autonomous flight of a UAV is a daunting task for any terrorist group, even were they to have all the necessary technical skills. It would require at least two years of determined effort and some level of outside or foreign assistance. Although Iraq produced its own UAVs based on indigenous design plans, they still depended greatly on foreign assistance for access to British piston engines and commercial autopilots, servomechanisms, and guidance software. Clearly, access to flight management systems not subject to export-control restrictions would save substantial development time and reduce the risk of catastrophic failure and detection. Devices like the ones acquired by the Iraqis cost in the neighborhood of $5,000, while providing complete autonomous operation, a GPS-aided autopilot, and automatic takeoff and landing. Still, the complexity and risks involved in achieving successful autonomous flight make the terrorist acquisition of a UAV a low-probability, but high-consequence threat, at present. But this assessment could change due to the rapid pace of technological change in all the technologies relevant to unmanned flight. Several new aerospace firms have emerged in the United States and other countries over the last five years to sell all the necessary components and services required for converting manned aircraft into UAVs. Controlling the sale of these technologies to legitimate parties and countries will grow increasingly problematic without national and multinational attention to the export control challenges these technologies present.
The First Line of Defense
By no means improbable and surely no longer unfamiliar, terrorist use of UAVs deserves a greater degree of attention than it receives today. The United States spends about $9 billion a year to defend against a few intercontinental ballistic missiles, the origin of which will always be evident. Yet, expenditures to improve detection and interception of low-flying cruise missiles and UAVs, whether emanating from offshore or domestic points of origin, pale by comparison. Given that it is impossible to conceive of an affordable and highly effective nationwide defense against these low-flying threats, it seems wise to pursue not only a more cost-effective but also perhaps the first and only real line of defense against these threats--improved nonproliferation policy.
To that end, because of the growing concern about terrorist use of kit airplanes and other civil aircraft used as makeshift and lethal UAVs, the United States introduced an "anti-terrorism" proposal within the 34-nation Wassenaar Arrangement (WA) in January 2003. Designed to place export control reviews on all equipment, systems, and specially designed components that would enable airplanes to be converted to UAVs, the U.S. proposal received little support within the WA due it its lack of specificity. At a March 2004 House Government Reform Committee hearing on ways of improving controls on cruise missiles and UAVs, the January 2003 "anti-terrorism" proposal to the WA was raised during the question and answer period. In response to a query, Lisa Bronson, director of the Defense Technology Security Administration and an undersecretary of defense, stated that the United States would reintroduce its proposal to control flight management within the WA. In light of the MTCR's more aggressive interest in assuring that its controlled technologies do not fall into terrorist hands, that regime as well should investigate closely how it might improve controls on UAV guidance and navigation technologies. By comparison with the WA, the MTCR possesses strong denial rules and no-undercut provisions (meaning if one member states denies a transfer, all must follow), making it the most effective multilateral mechanism to address the issue. The State Department's director of the Office of Chemical, Biological, and Missile Nonproliferation, Vann H. Van Diepen, indicated at the 2004 House hearing that his office planned to strengthen regulations dealing with missile guidance and propulsion. Although the United States alone cannot deal with the challenge of reducing terrorist access to technologies relevant to employing UAVs as weapons, it can exert its leadership role among key MTCR partners to convince the overall membership to reach consensus on this matter. Such a strengthening of export control regulations would make the prospect of transforming small manned aircraft into unmanned terror weapons far more difficult than it is today.
 Bradley Graham, "Terrorist Air Assault Caught the Pentagon Napping," International Herald Tribune, September 17, 2001, p. 1.
 Sandra I. Erwin, "Lack of a 'Single Integrated Air Picture' Hinders Commanders," National Defense Magazine, November 2001, www.nationaldefensemagazine.org.
 UAVs are broadly defined here, as they are in the MTCR regime, to include cruise missiles, target drones, and reconnaissance drones, capable of meeting the regime's range and payload parameters. More narrowly conceived, cruise missiles are always armed and are used just once, while UAVs, whether armed or not, are normally reusable, like manned aircraft. A terrorist, of course, would likely employ an armed or suicide UAV just once, like a cruise missile.
 David Pugliese, "Terrorist are training on hang-gliders, experts warn," Calgary Herald, March 26, 2004, p. 1.
 Daniel Benjamin and Steven Simon, "The Worse Defense," The New York Times, February 20, 2003, p. 31.
 The engineer is Bruce Simpson, www.interestingprojects.com.
 "DIY cruise missile thwarted," BBC News Online, December 9, 2003, at http://news.bbc.co.uk. According to Simpson, Iran was among a potential list of buyers of his detailed technical journals, which were offered on his website under a subscription option.
 See Lothar Ibrugger, Report of the NATO Parliamentary Assembly, Subcommittee on the Proliferation of Military Technology, Missile Defences and Weapons in Space, November 2004, http://natopa.ibicenter.net.
 HY-2 is the Chinese designation for what the West (or NATO) refers to as the Seersucker cruise missile. The HY-1, an earlier version, is referred to in the West as the Silkworm. All too frequently, Western media uses the Silkworm appellation for both the HY-1 and HY-2 anti-ship cruise missiles, both missiles are candidates for conversion to a land-attack cruise missile, www.fas.org.
 The UN-sponsored Law of the Sea Treaty specifies 12 nautical miles (or 22.2 km) as a nation's territorial waters, but in 1999 the President proclaimed that federal agencies would enforce American law up to twice the distance, or 39 km. According to Commander R.B. Watts, USCG, since 9/11 much closer scrutiny is made of suspect vessels wishing to enter U.S. waters and the Coast Guard routinely inspects such vessels before they are allowed to enter a U.S. port. Thus, standing considerably outside of U.S. waters would offer much greater prospect of avoiding detection prior to launch. See his article, "Fight Them Forward," in the monthly electronic journal produced by the Center for Contemporary Conflict at the Naval Postgraduate School, www.ccc.nps.navy.mil.
 All details discussed about the Jinin project are taken from the Comprehensive Report of the Special Advisor to the DCI on Iraq's WMD, Vol. 2, otherwise known as the "Duelfer Report," www.cia.gov.
 According to the Duelfer Report, Iraq's extended range HY-2 was flown to a distance of 168 km, but was adjusted to keep it within the UN-restricted range limit of 150 km because this program was declared by Iraq in 1996. One Iraqi engineer commented to the Survey Group that Iraq targeted Kuwait during the 2003 war with these missiles. However, no mention is made whether or not Iraq succeeded in acquiring and installing a modernized land-attack guidance and control system prior to use in 2003. For an analysis of the Seersucker's use in Operation Iraqi Freedom, see Dennis M. Gormley, "Missile Defence Myopia: Lessons from the Iraq War," Survival, vol. 45, no. 4 (Winter 2003-04), pp. 61-86.
 Bradley Graham, "Cruise Missile Threat Grows, Rumsfeld Says," Washington Post, August 18, 2002, p. A1.
 Marc Champion, David Cloud, and Anne Robbins, "At Davos, Powell Pushes Back Against Resistance Over Iraq," Wall Street Journal, January 27, 2003, www.lexis-nexis.com.
 This course of action would have required the installation of a suitable BW dispenser. The Iraqi Survey Group also reported that the Republican Guards had directed that some Al Musayara-20s be converted in weapons-carrying UAVs (conventional explosives). The Duelfer Report, www.cia.gov.
 Daniel Benjamin and Steven Simon, The Age of Sacred Terror (New York: Random House, 2002), pp. 203-205.
 Most other radioactive sources of concern, such as cobalt-60 and iridium-192, are solid materials. I am grateful to my former CNS colleague, Dr. Charles Ferguson, for bringing these points to my attention. See Charles Ferguson, Tahseen Kazi, and Judith Perera, "Commercial Radioactive Sources: Surveying the Security Risks," Occasional Paper No. 11, Center for Nonproliferation Studies, January 2003, http://cns.miis.edu.
 Richard A. Muller, "The Cropdusting Terrorist," Technology Review, March 11, 2002.
 "IDF Confirms Hizballah UAV Overflew Israeli Airspace for 'Very Short Time'," Jerusalem Voice of Israel Network B in Hebrew, April 11, 2005 [FBIS Translated Text].
 Spencer S. Hsu, "Plane That Caused Capitol Evacuation Nearly Shot Down," Washington Post, July 8, 2004, p. A1.
 Sara Kehaulani Goo, "Airspace Violations Prompt Rule Review," Washington Post, July 6, 2005, p. A6.
 This accounting was accomplished by Dr. Gregory DeSantis, a private consultant and colleague of the author, based on a review of the kit airplane literature during 2001.
 For an analysis see Eugene Miasnikov, Threat of Terrorism Using Unmanned Aerial Vehicles: Technical Aspects (Moscow: Center for Arms Control, Energy and Environmental Studies, Moscow Institute of Physics and Technology, 2005).
 Michael Gips, "A Remote Threat," Security Management Online, October 2002, www.securitymanagement.com; Gips also notes that one security expert, a former intelligence officer, Louis Mizell, has recorded 43 cases involving 14 terrorist groups where remote-control delivery systems were "either threatened, developed, or actually utilized."
 Duelfer Report, www.cia.gov; The Iraqi ambassador to Russia purchased GPS components for their UAV program from Russian technicians working at their homes, supposedly without the knowledge of the Russian government. Iraq also reportedly obtained four Micropilot MP2000 and two 3200VG autopilots from an Australia-based procurement agent.
 For an analysis of the cruise missile defense challenge, see Dennis M. Gormley, Dealing with the Threat of Cruise Missiles, Adelphi Paper 339 (Oxford: Oxford University Press, 2001).
 The WA attempts to achieve transparency and greater responsibility in transfers of conventional arms and dual-use goods and technologies (including UAVs with very short ranges, making it much more extensive than that of the Missile Technology Control Regime's range of at least 300km). See Dennis M. Gormley and Richard Speier, "Controlling Unmanned Air Vehicles: New Challenges," Nonproliferation Review, Summer 2003, pp. 66-79, for a review and analysis of recent nonproliferation initiatives, including the January 2003 U.S. antiterrorism proposal dealing with converting airplanes into UAVs.
 Robert Wall, "U.S. Intends to Enhance Controls Over Missile and UAV Technologies," Aviation Week & Space Technology, March 15, 2004, p. 25.
 Iraq's small UAV (known as Al Musayara-20) had the range (>500 km), payload (20 kg), guidance (GPS-aided), and autonomy needed to be an effective biological delivery means. Comprehensive Report of the Special Advisor to the Director of Central Intelligence on Iraq's Weapons of Mass Destruction, September 30, 2004.
 Iran is reported to be upgrading around 300 Chinese HY-2 Seersucker anti-ship cruise missiles by fitting them with a turbojet engine and new guidance systems.