The proliferation of uninhabited aerial vehicles (UAVs) has fundamentally altered the character of modern warfare, an observation repeatedly validated by their extensive, constantly evolving, and strikingly impactful use in recent conflicts. UAVs, however, vary considerably in type and characteristics, and this diversity is mirrored in the challenges they pose to defenders and, consequently, in the range of technologies, hardware, and tactics required to neutralise them.
At the top of the spectrum are larger, more exquisite types such as the MQ-9 Reaper and Bayraktar TB2, which fulfil surveillance and hunter-killer missions. Recent conflicts, however, have cast doubt on their survivability in contested airspace and against well-organised air defences. The more problematic UAV category is the one-way attack (OWA-UAVs), best exemplified by Iran's Shahed series. Also known as loitering munitions, these are unsophisticated airframes that can be produced quickly and in large numbers, thereby overwhelming and depleting a rival's air defences. Then at the bottom of the spectrum are First-Person View (FPV) drones: small, extremely low-cost, easily built flying devices, most commonly quadcopters. Marking a paradigm shift, Ukraine's Operation Spiderweb against Russia has demonstrated that traditional air-defence (AD) measures become largely irrelevant when FPV drones are smuggled to launch points near targets deep inside the victim's territory.
Traditionally, common techniques for neutralising UAV threats have been (i) jamming the positioning signals broadcast from satellites, and (ii) detecting and jamming the radiofrequency (RF) link between an air vehicle and its operator. Yet RF-spectrum countermeasures are being rapidly outpaced by relentless technological progress. Tethered drones, which trail a spool of fibre-optic cable to maintain a direct link with their operators, are immune to RF detection and jamming. Meanwhile, drones fitted with machine learning can now navigate without satellite signals or operator commands. AI-powered drones can also find, track, and strike targets with little or no human involvement.
Radar is another common instrument for detecting and cueing intruding UAVs However, long-range and always-scarce AD radars are not well-suited to finding low- and slow-flying drones. Moreover, large AD radars often find themselves in the crosshairs of combined drone-missile raids. The shift has consequently been toward shorter-range, less exquisite radars networked with a distributed array of passive (i.e., non-emitting) sensors, which are harder for intruders to locate and avoid. Foremost among passive detection techniques are acoustic sensors that listen for engine noise, though high false-alarm rates and the proliferation of silent, electric-powered UAVs pose challenges. Additionally, AI-enhanced electro-optic and thermal/infrared cameras are adept at automatically detecting airborne intruders. Still, they, too, suffer from limited range and adverse weather conditions such as rain, fog, and low cloud cover. The ideal solution is thus a hybrid sensor network combining multiple detection modes and techniques. But deploying such multi-node, multi-mode sensor networks is neither easy nor cheap.
Once intruders are detected, the next challenge is to neutralise them as quickly as possible. The choice embedded in all traditional AD doctrines; firing exquisite missiles such as Patriot or scrambling fighter jets armed with air-to-air missiles like AMRAAM is no longer workable. Most incoming OWA-UAVs cost as little as $20,000 apiece, whereas the missiles intercepting them are more expensive. This cost-exchange-ratio paradox, compounded by finite missile stocks, has rendered this category of AD unsustainable. It will take some years before a new generation of simpler and cheaper AD missiles reaches maturity and is deployed en masse. In the meantime, the U.S. has begun equipping its fighter jets with more affordable air-to-air missiles. The UAE has improvised by using helicopter gunships against Shaheeds. Ukraine's venerable transports are flying as drone hunters. And tests are underway for UAV motherships operating as anti-drone pickets, dropping small missiles and interceptor drones. However, defenders should be aware of the sizable risk of fratricide when dozens, even hundreds, of enemy drones crowd the same chaotic battlespace alongside friendly aircraft.
Another element of traditional AD is anti-aircraft artillery (AAA), which is particularly effective against low-flying OWA-UAVs. However, concerns over collateral damage from fragmenting projectiles rule out its use in densely populated areas or near critical infrastructure. AAA can still prove effective in desolate areas, in approach corridors frequented by intruding drones, or along exposed coastlines facing the sea. Also worth exploiting are uncrewed boats armed with rapid-fire guns, acting as long-endurance anti-drone pickets that push the defensive lines further out. Their additional benefit would be warding off unmanned surface vehicles that launch FPV drones against coastal installations.
Free of AAA's collateral-damage drawback, directed-energy (DE) weapons are frequently touted as a magical solution. The more visible of these are high-energy lasers, capable of unlimited shots, but only sequentially, striking one target at a time, a critical drawback against drone swarms. Their performance also declines sharply in rain and fog. Comparable to AAA in range, lasers are point-defence weapons, meaning large numbers must be deployed. A seven- to eight-figure price tag per unit implies that laser weapons are not cheap.
The second category of DE weapons, high-power microwaves (HPM), emit electromagnetic pulses that can turn off multiple drones at once by frying their circuitry and wiring at negligible per-shot cost. But their range is severely limited: a cylinder of energy no more than 1–2 kilometres long and roughly 150 metres wide. Adversaries, moreover, would find it easy to shield future drones against electromagnetic overload.
As the country targeted most extensively by drones, Ukraine has come up with a brilliant counter-drone solution: simple and cheap interceptor drones that ram into or detonate alongside enemy drones. 3D-printed, semi-autonomous interceptors sell for $1,000–3,000 apiece, shifting the cost-exchange ratio in the defender's favour. On the other hand, the expiration window in the struggle between intruding and intercepting drones is mind-bogglingly short: Ukrainian and Russian innovations reportedly become outdated every six weeks. One obvious implication is that, rather than stockpiling large quantities of interceptors in peacetime, states would do better by building the capacity in advance to mass-produce, procure, and continuously improve interceptors once hostilities break out.
An often-neglected dimension of counter-drone warfare concerns deception, concealment, and the hardening of potential targets. Nets and tension cables are examples of obstacles that disrupt predictable aerial access routes for drones, while camouflage and decoys offer practical, cost-effective protection. The war in Ukraine and recent conflicts in the Gulf have shown that there will always be flaws and never 100 per cent protection, underlining the need to revive fencing, hardening, and deception.
And lastly, pivotal to the success of any competing or complementary counter-drone approach is the availability of efficient command-and-control, communication, and battle management functions. At the heart of successful counter-drone operations lies seamless air-domain awareness and the ability to connect any sensor with any weapon in real time. Also needed are defensive elements linked to long-range strike assets, because purely defensive approaches will fail over time. The goal instead should be to reduce the number of attacks that must be intercepted in the first place.
In conflicts increasingly defined by scale, numbers, and speed, a related requisite is exploiting AI and its neural networks to track and control large numbers of drones, intercepts, frontlines, and resources. Most of the hardware and software is already there. Yet the bigger problem will be institutional, political, even cultural. Unclear command relationships created by drone defences, individual branches competing to preserve or expand their control and tutelage, old-fashioned procurement practices, and conflicts of interest between industries will likely prevail.
[1] This text was edited for English using AI tools, but all research, content and analysis are entirely the author’s own work.
