The Quiet Detector: Passive RF Sensing Changes Drone Defence Economics

To detect a drone, most systems broadcast energy and wait for the reflection. That broadcast is also a beacon. It tells every adversary with a receiver exactly where the sensor is, what type it is, and how to avoid it or destroy it. A growing number of defence buyers are choosing the opposite approach: detect nothing and therefore reveal nothing.

The passive RF market is expanding

R2 Wireless raised $5.3 million in a Series A round in January 2026, bringing its total to $13 million. The Austin-based company builds the Odin spectrum dominance platform, a passive system that monitors the full radio frequency spectrum to detect, classify, and geolocate transmissions from drones, cellular devices, Wi-Fi, and other wireless emitters. Odin transmits nothing. It learns the RF environment in real time and flags anomalies.

The company's technology was developed under Israel's high-threat environments before expanding into Europe and the United States. The company recently signed a formal agreement with a NATO member's Ministry of Defence, won the U.S. Army's xTech Disrupt competition in the counter-UAS category, and deployed its platform with the Department of War and the Department of Homeland Security. Each validates the same thesis: passive RF detection has moved from experimental to operational.

R2 is not alone. Guardian RF, a Y Combinator-backed startup, builds a distributed network of passive RF sensors called Scout that continuously observes drone activity without emitting. AISAR, a European company, classifies 37 drone types using a convolutional neural network on passive RF data alone. HADES Defence Systems in the Netherlands markets a passive RF sensor designed for NATO-compliant counter-UAS operations. Nokomis won a U.S. Army SBIR contract to develop a passive drone detection system based on unintended RF emissions.

None of these companies build radars. They all build listeners.

What active radar cannot do

Active radar has a fundamental constraint: it radiates. A radar that transmits to detect a drone also advertises its own location, frequency, and operating pattern. Adversaries equipped with electronic support measures can map active sensors, route around them, or destroy them. In contested environments, ground-based emitters (jammers, radar stations, drone control units) account for 85 percent of battlefield casualties, according to a NATO operational assessment published in July 2026.

Passive RF detection eliminates this exposure entirely. The sensor never transmits. It cannot be located, jammed, or targeted by an emitter-homing weapon. This makes it deployable in contested environments where active radars must be switched off or rotated to survive.

The second gap is scale. Small drones in class 1 and class 2 have radar cross-sections measured in hundredths of a square meter. A consumer quadcopter at 500 metres is invisible to most tactical radars. But every drone that is not flying autonomously on a pre-programmed route emits a control-link signal, a telemetry stream, or a video downlink. Passive RF sensors detect these emissions at ranges that often exceed the radar's detection distance for the same target.

AI turns RF noise into intelligence

The shift from detecting drones to identifying them is where AI changes the economics. Modern passive RF systems use deep-learning classifiers trained on spectrograms of known drone control protocols. The classifier reads the time-frequency pattern of a signal the way a facial recognition system reads a face. It matches against a library of known signatures and returns a model match within milliseconds.

AISAR's system runs a CNN that classifies 37 drone types from their RF fingerprints. The U.S. Army's SBIR programme has funded multiple efforts to build compact, low-power passive RF classifiers that can run on edge hardware. The academic literature (including a 2025 paper in MDPI Drones and a 2025 review in ScienceDirect) confirms that YOLO-based detection combined with CNN classifiers achieves high accuracy even under non-line-of-sight conditions.

The result is a sensor that can sit on a truck, a mast, or a drone, draw less power than a light bulb, emit nothing, and tell the operator exactly what kind of drone is approaching before the radar would see it.

Passive RF versus active radar in practice

The two approaches are not interchangeable. They cover different parts of the threat spectrum, and their operational profiles determine where each makes sense.

The pattern that emerges from the comparison is clear: passive RF covers the detection and classification layer at lower cost and zero electromagnetic footprint. Radar covers the coverage gap against RF-silent threats. A layered architecture that combines both is the direction most procurement programmes are taking. The question for defence buyers is no longer which technology to choose, but how to weight the two within a single sensor budget.

Procurement is shifting toward passive layers

The counter-UAS market is growing at 17.8 percent annually, but the growth is not uniform across sensor types. Active radar retains its role as the backbone of wide-area surveillance, but defence buyers are increasingly adding passive RF layers to their acquisition plans, driven by three factors.

First, the cost of passive sensors is an order of magnitude below tactical radar. A single Odin node costs a fraction of a ground-based radar set and covers a comparable detection radius for emitting drones. Second, passive sensors can be deployed without spectrum licensing, radio-frequency coordination, or the diplomatic clearances that active radar requires in allied operations. Third, the survivability argument has become operational doctrine. A sensor that cannot be detected forces the adversary to assume it is present, which constrains their behaviour even when the sensor is not there.

Honeywell's $1.9 billion acquisition of CAES in September 2024 illustrates the direction prime contractors expect the market to take. CAES specialised in RF and electronic warfare systems, including C-UAS sensing, and the acquisition price reflects investor conviction that passive RF will become a standard line item in defence procurement budgets, not a niche add-on.

EASA's mandate requiring C-UAS detection for all EU airports handling more than one million passengers per year adds a civilian demand driver that military spending alone would not create. The regulation covers passive detection technologies specifically, because airports operating active radar face interference with air-traffic control systems.

The fibre-optic blind spot

The emerging counter-argument is the rise of fibre-optic-controlled drones. A drone flying on a pre-programmed mission with a fibre spool instead of a radio link emits no RF signal at all. The U.S. Army has flagged this as a significant C-UAS challenge, and the only radar specifically tuned for fibre-optic drones is not expected to reach mass production until late 2027.

Passive RF cannot detect what does not radiate. This is a real limitation, and it is why layered detection architectures combining RF, radar, acoustic, and electro-optical sensors are becoming standard. The passive RF layer handles what emits. Radar and visual sensors handle what is silent. No single sensor covers the full threat spectrum.