Knowledge · RF-spectrum layer · Reviewed 2026-09
Spectrum Detection: How RF Sensors Find Drones That Broadcast Nothing
What RF-spectrum detection is, what it hears, what it misses, and how it chains with Remote ID receivers into a complete picture.
The principle: watch the airwaves, not the law
Remote ID receivers depend on cooperation — the aircraft must transmit its identity for a receiver to hear it. RF-spectrum detection drops that assumption entirely. Every drone, whatever it is, must emit radio to fly: a 2.4 GHz control link, a 5.8 GHz video downlink, or both. A spectrum sensor sweeps those bands continuously and flags the fingerprint of an airborne transmitter — even when the aircraft itself is trying to be invisible.
That independence is the layer’s value and its burden. It sees what RID cannot. But the 2.4 and 5.8 GHz bands are shared with Wi-Fi, Bluetooth and telemetry — so the sensor’s real skill is telling a drone’s signature apart from a laptop’s, and that is where design quality separates products.
The four technical capabilities that matter
Band coverage
Entry units watch 2.4 and 5.8 GHz. Serious ones add 900 MHz (long-range control links), 1.2/1.4 GHz (analogue video) and 5.1–5.2 GHz. Every band you do not cover is a door you leave open — the right question is which bands your threat actually uses, not which list is longest.
Protocol decoding
Some sensors decode DJI’s proprietary Drone-ID directly from the RF signal — recovering aircraft position without any Remote ID broadcast. Others fingerprint the transmitter: recognising the RF signature of a specific drone family. Decoding gives identity; fingerprinting gives detection. They are not the same promise.
Direction finding (DF)
DF hardware estimates the bearing to the transmitter. Vendors commonly claim ±5° accuracy; real-world figures depend on multipath and antenna quality — which is why DF claims deserve the same scepticism as range claims. Two DF sensors can triangulate a position; one only gives you a bearing to walk toward.
Feature libraries and updates
Signature-based detection depends on a maintained library of drone RF fingerprints. Ask who updates it, how often, for how long, and what happens when a new aircraft appears. A stale library degrades quietly — the sensor keeps working while its list of known threats ages.
What RF-spectrum detection cannot do
- It cannot reliably tell you who is flying — identity is the RID layer’s job
- It raises false alarms in Wi-Fi-dense environments, and every alert costs operator attention
- It does not stop, jam or ground an aircraft — and we do not advise on measures that do
- Feature libraries need ongoing updates; a neglected sensor degrades silently
This is why we evaluate it as a chain, not a replacement: the RF layer says something is flying, the RID layer says whose it is and where it launched. If your threat model includes modified or FPV aircraft, the combination is the honest answer; if it does not, RID alone is the cheaper and quieter one.
Costs and the open-source route
Commercial RF-spectrum units run $3,500–15,000 ◐ vendor-reported (checked 2026-09). For technical teams, the open-source route is real: WarDragon and AryaOS approximate the layer on $300–500 of SDR hardware — with Linux and SDR skills required, no support line, and documentation you maintain yourself. We document these builds because understanding the technology makes you a better buyer of it.
Band coverage, DF accuracy and false-alarm figures on the candidate table are vendor-reported with check dates — not measured by us. Our two-layer field validation is in progress.