FUTURECRAFT | TECHNOLOGY & MARKETS
A Baaj Akh unit hums through the wheat fields outside Tarn Taran most nights now, its jammer sweeping the dark for rotor noise. An Indrajaal Ranger idles at a forward post further south, its mast up, its AI core watching four kilometres of sky in every direction. Between them and the dozen systems like them now strung along the Punjab and Rajasthan border, India has built the densest counter-drone shield it has ever fielded, and Union Home Minister Amit Shah has promised more within six months.
The drones keep coming anyway. Security agencies intercepted 294 in 2024 and 287 in 2025, a number that has barely moved even as the hardware protecting the border has gotten sharper, per Indian Express reporting on the western sector. Pakistan-based handlers have upgraded their fleets too, switching to satellite-linked, anti-jamming models that exploit a narrow window between 10 pm and 4 am, when 12 to 15 drones can cross in a single hour.
The uncomfortable part is not that the shield has gaps. It is that the shield is telling everyone, in real time, exactly where those gaps are.
THE SIX-MONTH CLOCK
Shah made his promise at a Border Security Force outpost in Bikaner on May 26, laying out a timeline to install anti-drone systems across the international border to stop narcotics and weapons moving in on quadcopters. The urgency is not rhetorical. Punjab’s 462-kilometre border fence, built in the late 1980s and 1990s to stop people and vehicles, was never designed to stop anything that flies, and traffickers know it.
What has changed is the sophistication on the other side. Drones intercepted along the border increasingly carry anti-jamming firmware and satellite-linked navigation, according to the same reporting, which means they can shrug off a jamming signal and still find their way home. Fail-safe return-to-home settings let a drone that loses its link simply retrace its path, denying the ground crew both the cargo and the wreckage. This looks less like an amateur smuggling run and more like a supply chain that has spent years iterating against India’s defences and learning precisely which counter-measures to route around.
TWO MACHINES, ONE INSTINCT
The state’s answer has been to build louder, not quieter. Punjab’s Baaj Akh, Punjabi for eagle eye, went live in Tarn Taran on August 9, 2025, under Chief Minister Bhagwant Mann, after the state government sanctioned nine units in May that year. It detects, tracks, jams and neutralises targets out to 8 to 10 kilometres, and Punjab described itself as the first state to deploy anti-drone systems at an international border.
A year later, the Army placed a ₹155 crore order for the Indrajaal Ranger, a mobile anti-drone patrol vehicle built by Hyderabad-based Indrajaal Autonomous Defense Systems with partner Sigma Advanced Systems. A separate ₹10 crore order came in for urban policing and VVIP convoys. The Ranger folds RF jamming, GNSS spoofing, a mid-flight cyber takeover capability and a physical kill switch into one vehicle, run by an AI core called SkyOS that the company says coordinates a response within seconds of a detection.
Every one of these systems shares a design assumption: that the way to stop a drone is to reach out and touch it, electronically or otherwise. To do that, the machine has to switch something on.
WHY SWITCHING IT ON GIVES YOU AWAY
Radar and jamming equipment share a property that predates drones by decades. To function, they transmit, and transmission has a signature. Specialists in radio-frequency detection describe the asymmetry bluntly: active systems broadcast their own presence to anyone equipped to listen, while a passive receiver network can locate that transmitter without emitting anything of its own, using nothing more than the time it takes a signal to reach several receivers at slightly different moments. Four receivers and a stopwatch, in effect, are enough to fix a jammer’s position in three dimensions.
This is the operating principle behind every Baaj Akh unit and every Indrajaal Ranger the moment its jammer keys up along the border. A mesh of these systems, spaced to cover Punjab’s smuggling corridors, produces a live, continuously refreshed map of exactly where India’s detection coverage sits and, by subtraction, where it does not. Handlers who already study patrol shift changes and fly in narrow overnight windows do not need much more sophistication to start listening instead of watching. The intelligence discipline required is smaller than the one they have already built to defeat the jamming itself.
THE CASE FOR STAYING LOUD
None of this argues for switching the jammers off. A passive network can tell you where a threat is transmitting from, but it cannot force a drone down or scramble its navigation, and interdiction still needs an active layer somewhere in the loop. The steelman here is straightforward: Punjab’s own numbers, while flat in aggregate, do not capture how many high-value shipments were interrupted rather than delivered, and even a partial deterrent has value against a corridor this porous.
The real gap is that India already has the architecture to fix this and is not spending against it. Akashteer, the Army’s indigenous, AI-driven air defence control network, fused radar and interceptor data into a single real-time picture during Operation Sindoor in May 2025 and is built precisely to absorb more sensor feeds, including passive ones. Procurement money, though, keeps going to vehicles that can be photographed making a kill, the ₹155 crore Ranger order and the ₹10 crore urban add-on among them, rather than to the quieter receivers that would tell commanders where their own coverage is exposed before an adversary works it out first.
BOTTOM LINE
Shah’s six-month clock runs out around November. What gets installed in that window will almost certainly look like more Baaj Akh units and more Rangers, because those are the systems that generate a press release and a photograph of a downed quadcopter. The 287-drone plateau will not move until border security budgets treat a receiver that says nothing as seriously as a jammer that announces itself to the entire corridor. Until then, every new system switched on to close the sky is also, quietly, redrawing the map for whoever is patient enough to just listen.
(The Author studies Computer Science and Artificial Intelligence at Rutgers University, New Jersey, USA. He is interested in emerging technologies and innovation, and can be reached on LinkedIn at @arssh-kumar14)
