How Drone Detection Works: RF vs Radar vs EO/IR vs Acoustic
Updated 16 October 2026 · BlackByrne · ~6 min read
Drone detection relies on four sensor types. RF scanners listen for control and video links. Radar tracks returns from the airframe. EO/IR cameras see and classify the drone. Acoustic arrays hear its rotors. Each has blind spots, so fielded systems fuse them into one track picture rather than trusting any one alone.
The four detection methods
Radio frequency (RF)
RF sensors scan the spectrum for the control link, video downlink, or Remote ID broadcast between a drone and its operator. Many can identify the make and protocol and estimate the direction to both the drone and the pilot. RF is passive, needs no transmitter of its own, and often gives the earliest warning. Its weakness is simple: a drone that does not emit cannot be heard. Autonomous drones flying on inertial or optical navigation, and fiber-optic tethered drones with no radio link, are invisible to RF. Crowded spectrum in cities also raises false alarms.
Radar
Radar transmits radio energy and measures the reflection from the airframe, producing range, bearing, and a track. It covers wide areas, works day and night, and works through most weather. The problem with small drones is physics: the radar return of a consumer quadcopter is close to that of a bird, which drives false alarms, and urban multipath and clutter can mask a low, slow target. Radar also emits, so it can be detected.
Electro-optical and infrared (EO/IR)
EO/IR uses daylight and thermal cameras to see the target. It is the layer that confirms what the other sensors found and classifies it, which most rules of engagement require before any response. Its field of view is narrow, so it works best when another sensor cues it onto a bearing. Daylight cameras struggle at night and in haze; thermal helps but still degrades in rain and fog.
Acoustic
Acoustic arrays detect the distinctive sound of rotors and motors. They are passive, inexpensive, and can hear a drone that emits no radio signal, which makes them useful against autonomous and tethered threats. Range is short, typically a few hundred meters, and performance falls in wind and ambient noise. Acoustic is usually a close-in cueing and gap-filling layer. Ukraine fielded a national acoustic sensor network reported to cost under 5 million dollars.
Side-by-side comparison
| Method | How it works | Strengths | Weaknesses | Works best as |
|---|---|---|---|---|
| RF | Detects and decodes control, video & Remote ID links | Passive; long warning range; can identify model & operator direction | Blind to non-emitting, autonomous, or fiber-optic drones; spectrum clutter | Early warning and attribution |
| Radar | Transmits RF, measures the reflected return | All-weather; wide area; gives range and track | Small-drone return resembles a bird; urban clutter; emits | Wide-area surveillance and tracking |
| EO/IR | Daylight and thermal cameras image the target | Visual identification and classification; thermal works at night | Narrow field of view; degraded by weather and darkness; needs cueing | Positive identification before response |
| Acoustic | Microphone arrays detect rotor and motor noise | Passive; low cost; hears non-emitting drones | Short range; defeated by wind and ambient noise | Close-in cueing and gap-filling |
Why no single method is enough
Every method has a blind spot that another method covers. RF misses silent drones; acoustic and radar can still find them. Radar confuses drones with birds; EO/IR and RF resolve the ambiguity. EO/IR sees a narrow slice; radar and RF tell it where to look. Acoustic reaches only a short distance; radar and RF extend the range. An attacker who defeats one sensor rarely defeats all four at once. That is why detection is a layered problem, not a single-sensor purchase.
UAS size matters: the DoD Groups 1 to 5
The US Department of Defense sorts unmanned aircraft into five groups by weight, altitude, and speed (JP 3-30). Size shapes which sensor works:
- Group 1: 0 to 20 lb, normally under 1,200 ft AGL, under 100 knots. Small and quiet; hardest to detect.
- Group 2: 21 to 55 lb, under 3,500 ft, under 250 knots.
- Group 3: under 1,320 lb, under 18,000 ft MSL, under 250 knots.
- Group 4: over 1,320 lb, normally under 18,000 ft MSL.
- Group 5: over 1,320 lb, normally above 18,000 ft MSL.
A system placed against a larger group is placed in the higher group if any one attribute exceeds the lower limit. The small Group 1 and 2 drones are the ones that most defeat single sensors, which is why fusion matters most there.
Sensor fusion: one picture
Fusion combines the sensors into a single track with one identity and one location, instead of four separate alarms an operator must reconcile. A common pattern lets RF and acoustic give early warning, radar build and hold the wide-area track, and EO/IR confirm and classify before any decision. Correlating detections across sensors also cuts the false alarms that wear operators down. Fusion is the foundation of counter-UAS: detect, then decide, then act on one clear picture.
FAQ
What is the best single drone detection method?
There is no single best method. RF, radar, EO/IR, and acoustic each have a blind spot another covers. Fielded systems combine them, because the drones that defeat one sensor are usually caught by another.
Can you detect a drone that is not transmitting?
Yes, but not with RF. A silent drone flying on inertial or optical navigation, or on a fiber-optic tether, is found by radar, EO/IR, or acoustic sensors, which do not depend on the drone emitting a radio signal.
Why is radar unreliable against small drones?
A small quadcopter reflects about as much radar energy as a bird. That similarity produces false alarms, and urban clutter and multipath can hide a low, slow drone. Radar still gives valuable range and track, which is why it is fused with other sensors rather than used alone.
What are the DoD UAS groups?
Groups 1 to 5 classify unmanned aircraft by weight, altitude, and speed. Group 1 is under 20 lb and Group 2 is 21 to 55 lb; Groups 3 to 5 are heavier and faster, up to large aircraft above 18,000 ft. Smaller groups are the hardest to detect.
DaggerOS fuses RF, radar, EO/IR, and acoustic detections into one track picture for counter-UAS operators. See DaggerOS counter-UAS for how the sensors are combined.
Sources: US DoD UAS group classification (JP 3-30); FAA Aeronautical Information publications on large UAS; Fraunhofer counter-UAS research; Atlantic Council / Aviation Today on drone detection limits; Counter-UAS 101, acoustic drone detection.