Micro-Doppler signatures
Drone rotors spin at 5,000–10,000 rpm. They create high-frequency Doppler sidebands around the main return — a 'JEM' (jet-engine modulation) signature. Birds flap their wings 2–10 times per second, producing a very different micro-Doppler pattern. With enough pulse coherence (typically 128–256 pulses), the classifier can distinguish quadcopter, fixed-wing drone and bird with 95%+ accuracy.
Sensor fusion
Pure radar is rarely enough. Modern C-UAS systems fuse radar with RF detection (sniffing the drone's control link), electro-optical cameras with computer vision and sometimes acoustic sensors. Radar provides the cue ('something at 247° azimuth, 1.4 km'), the camera confirms drone vs bird and the RF subsystem identifies the make and model.
Defeat mechanisms
Once classified, the system can jam the control link, spoof the GPS or — in military contexts — engage with kinetic effectors. Stadiums and airports prefer non-kinetic defeat to avoid the drone falling on a crowd. Systems like AUDS, DroneShield and Anduril Sentry are deployed at major airports.
Swarms
The next problem is swarms. A single radar can track dozens of drones, but engaging 50 simultaneously exceeds most jammers' bandwidth. High-energy lasers and high-power microwaves are emerging as the only practical defeat against true swarms.
Frequency Selection and Range-Resolution Trade-offs
In the C-UAS domain, frequency selection is a critical engineering trade-off between atmospheric attenuation and spatial resolution. High-frequency bands like X-band (8–12 GHz) and Ku-band (12–18 GHz) are the industry standard for tactical drone detection. These wavelengths are short enough to resolve small targets with a low Radar Cross Section (RCS) while maintaining high angular accuracy. However, higher frequencies suffer from greater path loss in high humidity or heavy rain, which can reduce detection ranges from 5 kilometers to under 2 kilometers. Consequently, most permanent installations utilize tiered architectures where S-band radar provides wide-area volume search, and X-band sensors are queried only once a potential track is established.
The range resolution—the ability to distinguish between two objects close together—is governed by the signal bandwidth. To distinguish a drone from a nearby bird or a second platform in a multi-rotor configuration, systems require bandwidths exceeding 50 MHz. Modern Gallium Nitride (GaN) based Active Electronically Scanned Arrays (AESA) allow for high power density and rapid beam steering without moving parts. This allows the radar to maintain 'track-while-scan' operations, simultaneously updating the position of 50 or more targets while continuing to sweep the horizon for new arrivals. This technical leap was instrumental during the 2018 Gatwick Airport incident, where legacy systems failed to maintain continuous tracks on intermittent targets.
False Alarm Suppression and Ground Clutter
Detecting a drone at low altitudes requires extracting extremely weak signals from massive ground clutter. Trees, fences, and moving vehicles can have an RCS several orders of magnitude larger than a 1 kg quadcopter. To mitigate this, counter-drone radars use high-dynamic-range receivers and sophisticated Moving Target Indicator (MTI) filters. These filters ignore stationary objects but must be carefully tuned to avoid filtering out drones that have entered a hover state. If a drone stops moving relative to the radar, its primary Doppler shift drops to zero, making it invisible to standard MTI algorithms—a common failure point in early generation security radars.
To solve the 'hover' problem, modern processors analyze the residual Micro-Doppler from the spinning propellers even when the drone's fuselage is stationary. This requires a high Pulse Repetition Frequency (PRF) to avoid Doppler ambiguity. Advanced systems now integrate Constant False Alarm Rate (CFAR) algorithms that dynamically adjust detection thresholds based on local environmental noise. This prevents the system from being overwhelmed by false detections caused by wind-blown foliage or insect swarms. In critical infrastructure protection, maintaining a low false-alarm rate is as important as detection probability, as constant false triggers eventually lead operators to ignore or disable the system entirely.
Phase-Interferometric Resolution for Low-Altitudes
Traditional pulsed-Doppler systems often struggle with multipath propagation when drones fly below 50 meters. At these altitudes, the signal bounces off the ground, creating 'ghost' images that fluctuate wildly. To combat this, modern 3D AESA (Active Electronically Scanned Array) radars utilize phase-interferometry. By comparing the phase difference of the return signal across several receiver sub-apertures, the system can calculate a highly precise elevation angle. This allows the radar to distinguish a drone hovering just above the tree line from the stationary clutter of the foliage itself, even when the drone's radial velocity is near zero.
Precision is further enhanced through the use of LFM (Linear Frequency Modulation) chirps. By sweeping a wide bandwidth, typically 50–200 MHz, the radar achieves a range resolution of less than 1.5 meters. This spatial granularity is essential for resolving individual drones within a tightly packed formation. While older legacy systems might see a swarm as a single large blob, high-resolution wideband radar can count the distinct airframes, providing the command-and-control system with an accurate count of threats for prioritized engagement or jamming resource allocation.
The Carbon-Fiber Blind Spot
A common misconception is that all drones are equally detectable; however, the material composition of the airframe significantly alters the Radar Cross Section (RCS). While plastic quadcopters reflect signals largely from their internal electronics and motors, carbon-fiber reinforced polymer (CFRP) frames are partially conductive. At X-band frequencies (8–12 GHz), carbon fiber can behave like a lossy dielectric, absorbing some incident energy rather than reflecting it. This reduces the detection range by up to 30% compared to a similarly sized metallic or high-density plastic airframe, necessitating higher peak power from the transmitter.
The shift toward stealth-optimized geometry in consumer-grade DIY racing drones further complicates detection. Simple modifications, like canting the electronics housing or using non-orthogonal surfaces, can deflect radar energy away from the receiver’s line of sight. To counter this 'low-observable' trend in the hobbyist sector, modern counter-drone radars are shifting toward S-band or dual-band operations. By utilizing longer wavelengths (around 10 cm), the system experiences less atmospheric attenuation and can better detect the structural resonance of the airframe, ensuring the target remains visible even when its RCS is minimized.