We should get a counter-UAS system soon

September made this argument for me. In three weeks, drone swarms shut down a Russian oil refinery. An Iranian retaliation folded drones and ballistic missiles into one Gulf-wide barrage. Reports from West Asia and the Black Sea pointed to unmanned systems reaching energy infrastructure and merchant shipping, well clear of any front line. None of this is abstract for India; it is the same threat that reached Indian skies during Operation Sindoor last year, now arriving elsewhere at a scale that should settle, once and for all, whether counter-UAS is a line item or a doctrine.

Start with the night of September 21-22. Russia sent 212 drones and a mix of missiles at Ukraine in a single overnight barrage. Ukrainian air defence destroyed or jammed 186 of them, by the Ukrainian Air Force’s own count. The rest got through — killing at least three people in Dnipro, tearing up rail infrastructure, cutting power to tens of thousands of homes. Stopping 88 per cent of an attack and still losing lives is not a technology failure. It is an arithmetic one. There is only so much a finite stock of expensive interceptors can do against an adversary who can put hundreds of cheap aircraft into the sky in one night.

Days earlier, Ukraine had run the same numbers the other way — its largest drone strike package to date, officials said, against a major Russian oil refinery and its logistics hubs. Most of the incoming drones were reportedly intercepted.

The refinery shut down anyway. The lesson cuts both ways: whichever side of the border you sit on, if you cannot sustain your interception rate against sheer volume, you concede the target.

The same pattern showed up in a completely different theatre 2,500 kilometre (km) away.

The ceasefire between the United States (US) and Iran, already fragile since earlier in the year, broke down again in early September after a US strike was alleged to have hit a wedding gathering, killing civilians. Iran did not answer with one retaliatory strike. It answered with missiles and drones together, aimed at American and allied positions across Jordan, Bahrain, Kuwait and Iraq, in a single night. The drone, once a niche reconnaissance tool, now rides alongside the ballistic missile in a state’s retaliation toolkit — not instead of it.

There was more through the month — Houthi drone and missile activity aimed at Saudi energy infrastructure, Ukrainian sea drones striking tankers in the Black Sea, far from any declared front. But the shape of the month is not really in dispute. Drones are not staying inside declared battlefields anymore. They are reaching pipelines, refineries, ports, merchant ships as part of coordinated,multidomain packages, not stray nuisance raids.

Strip away the geography and every one of these theatres has the same thing in common. The attacker never sent one weapon type. Missiles and drones, or drones and decoys, together — because a mixed package stresses a defender’s decision cycle in a way a single weapon never does. A radar operator tuned to catch missiles is not tuned to catch quadcopters skimming below the clutter line.

Ask him to watch for both at once and his system either has a fused picture to cope, or it does not. Even Ukraine, with years of combat-tested electronic warfare behind it, still loses roughly 12 per cent of an overnight barrage to the seams in its own coverage. A country starting from zero should not assume it will do better on its first real night. None of this is someone else’s problem for India to read about at a distance. It is a preview of what a coordinated drone-and-missile package aimed at Jamnagar or Paradip, at ports along the western seaboard, at forward airbases on two live fronts, would actually look like. And the common thread running through every one of these incidents is this: no single device, however good the jammer or capable the radar, stopped a determined, layered attack on its own. What stopped part of each attack was a network acting as one system. What let the rest through was that system’s limits under volume and pressure. Neither outcome had much to do with any one piece of architecture.

India’s own experience last year, during Operation Sindoor, delivered a similar split verdict. Pakistan sent several hundred drones — Turkish-origin Bayraktars, locally assembled micro-drones — across roughly three dozen locations in one mass incursion. Akashteer, India’s air defence control network, tied into the Integrated Air Command and Control System (IACCS), caught most of it and struck back precisely. By most independent and official accounts, that was networked defence working the way it is supposed to. It is also, understandably, the success that has shaped how confidently India has approached everything it has bought since. 

What that confidence has also done, less usefully, is harden a procurement instinct. Since Operation Sindoor, India has moved fast on individual pieces: the man-portable T-Shul Beam jammer, demonstrated at the Pokhran range; a planned induction of roughly 30,000 drones across a 35-km border envelope, with the army already running most drone and anti-drone work in that zone; specialised Baaz Battalions, raised under 11 Corps in Jalandhar, built to pull drone use out of fragmented, unit-level hands and into a centralised structure. Each of these is a sound buy on its own terms. Add them up and they still do not make a system.

To be fair, some of this is being noticed and worked on. Defence technology company, IG Defence, which makes the T-Shul Beam, has started integrating it with GRID, a battle management layer meant to fold counter-drone capability into one shared picture — a vendor, of all people, admitting that a jammer outside a network is only doing half its job. The same company’s KAL loitering munition is being built alongside that same networked logic rather than as a standalone strike round. Indrajaal, a separate venture, has fielded a mobile anti-drone patrol vehicle built to move with a threat instead of waiting for one to arrive. All useful. None of it replaces a single doctrinal owner deciding how every piece talks to every other.

Even the paperwork is starting to catch up with a request for proposal for 83 tracked carrier air defence tracked vehicles asking for Akashteer, drone detection, drone interdiction and drone-kill systems to be built onto a single platform from the start — tied into the IACCS and the navy’s Trigun network, rather than bolted together once separate contracts land. Small detail. Telling one. India’s own procurement machinery is beginning to think in systems, not serials.

Here is what should worry India’s planners more than any hardware gap. Army electronic warfare (EW) regiments sit under signals and intelligence commands, built for communication intercepts and battlefield jamming. The air force runs its own EW squadrons, built mainly to protect aircraft. The two rarely share a sensor-to-shooter loop. Add a handful of ageing Phalcon and Netra aircraft, rarely all mission-ready at once, and you get real gaps in tracking low-flying threats over any meaningful stretch of contested sky. In a saturation attack, where EW’s whole value lies in disrupting a swarm before it ever reaches a kinetic interceptor, that institutional split is not paperwork. It is exactly the seam an adversary’s saturation doctrine is designed to find.

India’s own defence establishment is starting to say this out loud, too. General Anil Chauhan (retired), former chief of defence staff, had publicly called for a holistic counter-UAS system rather than a pile of devices,warning against exactly the fragmentation this piece is describing. The army’s own recent request for a joint development partner on a sovereign swarm algorithm made the same point in procurement language: a common architecture of its own, not another vendor-locked box. The instinct exists. It just has not caught up yet with the pace at which individual devices keep landing.

And this is not only a military problem. India’s airports, power substations, refineries, prisons — they share the same weakness the Saudi pipeline strikes and the Black Sea tanker attacks exposed elsewhere this month. Security that ends at the perimetre fence, built on the old assumption that a threat arrives on the ground, not from above. A jammer at the main gate does nothing for a facility with open approach paths on every side. Every refinery, port and power installation on the critical-infrastructure list needs the same layered, threat-assessed cover the army is now building for its border formations coordinated with whichever ministry actually owns the site, not left for each facility to figure out alone.

Counter-drone technology itself is not exotic or new. It rests on two simple jobs: detect, then interdict. Detection draws on radar tuned to small, slow targets; radio-frequency sensors that pick up a drone’s control link; electro-optical and infrared cameras for visual confirmation; and acoustic sensors that recognise the particular whine of a drone motor. No single sensor is reliable on its own, which is why serious systems fuse two or three of these together. Interdiction follows the same logic of layered options — soft-kill methods such as RF jamming, global positioning system jamming and signal spoofing that sever a drone’s link to its operator; and hard-kill methods such as lasers, nets, interceptor projectiles and directed-energy weapons for whatever soft-kill does not stop. None of this is classified or futuristic. What is missing in India’s case is not the list of technologies. It is the wiring between them.

Here is one way to see the difference clearly. A detailed 2025 study by the Takshashila Institution lays out six functions a counter-UAS architecture actually needs to perform: prevention, deterrence, denial, detection, interruption, destruction. A jammer or an RF gun lives entirely inside the last two. It does nothing for the other four — the intelligence that flags a threat before it launches, the deterrence that makes non-compliance costly, the passive measures that deny a target its vulnerability, the networked grid that has to exist before any interruption decision gets made at all. Buy more interruption and destruction gear while ignoring the first four functions, and you have reinforced the goalkeeper while leaving the rest of the defence unmarked. 

India already has a layered air defence construct in place, and it is worth being clear about that rather than pretending the country is starting from nothing. The first layer, closest to the threat, is meant to be counter-drone systems and man- portable air defence; the second, short-range surface-to-air missiles and point air defence; the third, medium-range missiles; the fourth and outermost, long-range systems, with Akashteer and the IACCS meant to fuse all four into one recognised air picture. The layers exist on paper and increasingly in the field. What India has not yet proven at scale is that a detection at the first layer reliably and quickly escalates to the right response at whichever layer is needed, every time, without a human bottleneck slowing it by passing on the responsibility.

The same Takshashila study sketches what the grid should look like on the ground along the international border and the Line of Control — where the threat geometry is linear, it proposes a tiered grid — systems layered in depth, density matched to actual threat trends, plugged rearward into the air defence network so a detection at the forward edge can escalate smoothly to a weapon further back. Around cities and critical infrastructure, where the geometry looks more like a honeycomb than a line, it proposes a mesh of overlapping zones around each vulnerable site, with one standard procedure across whichever agency happens to be guarding that particular spot. Neither model survives as a pile of standalone boxes. Both need a lateral plug-in between air traffic control, the unmanned traffic management system, the counter-UAS grid and the air defence network — all of it synthesised into one real-time picture at the IACCS.

Counter-drone capability is a system-of-systems — detection, identification, decision-making, neutralisation — all run through a time-critical loop where automation and pre-authorised rules of engagement matter more than any sensor’s range or any jammer’s power. I have stood in forward command posts and operational rooms watching air photos take too long to sort a genuine threat out of clutter. The fix was never a better radar. It was pre-cleared rules of engagement, and a networked picture that lets the man on the ground act on what he actually sees instead of waiting on a chain of command that a fast-moving swarm has no patience for.

The cost side of this makes the case even sharper. Independent estimates put it bluntly: intercepting a swarm of a few hundred drones with kinetic interceptors alone — Akash, Barak-8, Spyder could run into tens of millions of dollars in a single engagement, against an attacker spending a fraction of that to generate the same volume of threats. No missile inventory survives that exchange rate for long. However large it is.

The only answer that holds up over time is soft-kill first — electronic disruption handling the bulk of a swarm at near-zero marginal cost, kinetic interceptors saved for the leakers and the genuinely high-value threats. India already has real pieces of this. 

RayStrike-9, an indigenous 3-9 kilowatt laser from Paras Defence and the Defence Research and Development Organisation, is built on exactly that infinite-magazine logic. The navy has contracted Bengaluru-based defence technology firm Tonbo Imaging to build a high-power microwave weapon under its ADITI 3.0 scheme. Both are genuine soft-kill assets. Neither means much until it is wired into the same decision loop as everything else which is, again, a question of sequencing and command authority.

Recent Indian defence research has a useful shorthand for this: detect, survive, sustain. Detect through a distributed mesh of cheap sensors rather than a handful of expensive radars. Survive through redundant, mobile nodes, so knocking out one sensor does not blind the whole grid. Sustain through pre-stocked ammunition and modular resupply, so a defender is not empty three hours into a raid that runs all night. None of the three is a piece of hardware. Each is a standing decision about how the network gets built, paid for and commanded.

That, again, is the whole difference between an equipment list and an architecture.

None of this argues against buying good equipment. India needs T-SHUL BEAMs, low-level light-weight radars, Baaz Battalions, and the sovereign swarm capability it is chasing. What it needs alongside all of that, on the same timeline, is the doctrinal decision that ties the pieces together. A mandated, time-bound push to take Akashteer-IACCS integration to every theatre command, not just a handful of demonstration sites. One institutional home for electronic warfare and air defence, instead of two services running separate stacks that barely speak. A soft-kill-first sequence written into standing rules of engagement, so a forward unit facing a swarm at dusk is not waiting on a decision someone should have pre-authorised months ago. India has already shown it can move fast when it wants to. The Baaz Battalions went from concept to a raised unit along the Pakistan border within months of Operation Sindoor. Counter-UAS deserves that same urgency. The same institutional seriousness.

An architecture is the doctrine that tells every formation, every service, how to fight together when the sky over a forward post fills with more targets than any single device was ever built to handle. Give credit where it belongs. The Baaz Battalions, the CADET programme, the sovereign swarm capability, the T-SHUL BEAMs and RayStrikes now under trial, the training grounds at Pokhran turning out troops who can actually fly and fight these systems — none of this happened by accident. The government and the services have moved with a pace and a seriousness that a decade ago would have been hard to imagine. India has the technology base, the industrial capacity. Increasingly, the institutional appetite too.

What none of it adds up to yet, on its own, is a doctrine. Good troops with good equipment still need one command architecture to fight as a single system, and one national security doctrine that states, in writing, how counter-UAS sits alongside air defence rather than trailing behind it as an afterthought. That is the piece still missing — not effort, not intent, but the architectural decision that ties every good instinct on display this year into one coherent whole. I would rather India spend the next year arguing over command protocols and integration timelines than spend a single night finding out what it costs to get this wrong — the way Kyiv, Dnipro and a wedding party in Iran found out, this September. Four different theatres, one month, and each one showed exactly what sits on the other side of this decision. The devices are being bought. The doctrine that gives them meaning is still being written. For India to become Viksit Bharat by 2047, it will first have to become surakshit bharat and that is not a hardware question. It is a question of governance.