VolAnti: Open-Source Acoustic Drone Detection That Works Without a Radio Signal
Open-source acoustic drone detection. It hears the propellers, not the radio, so it works against fibre-optic FPV aircraft that emit no signal at all.
At a glance
- What is it?
- VolAnti is an open-source hardware and firmware project for detecting multirotor drones by listening for the harmonic comb their propellers produce, not by scanning the radio spectrum. A 91 mm box with four MEMS microphones, an ESP32-S3, an e-paper display, and a LoRa radio, it alerts locally and notifies other units on the network within 0.23 seconds of detection.
- Who is it for?
- VolAnti fits schools, farms, and small security operations that need a low-cost local tripwire for drones in the last few hundred metres, especially against fibre-optic FPV aircraft that emit no radio signal. It is not a replacement for radar or a full site security system.
- Can I use it commercially?
- Check first. The repository uses a licence we do not classify automatically, so read its LICENSE file before any commercial use.
- Is it still maintained?
- Yes. The repository last received commits 10 days ago.
- What is it written in?
- Mainly C, according to GitHub's language statistics.
Answers come from the project's GitHub data, last synced on September 30, 2026, and from our analysis. They are not legal advice.
Editorial analysis
The Problem: Fibre-Optic FPV Drones and Silent Radio Spectra
Nearly all commercial drone detection works by scanning the radio spectrum for the control link between pilot and aircraft. That approach has one failure mode: the drone must transmit. A fibre-optic FPV drone trails a glass fibre back to its operator. Video comes down the fibre, control goes up it, and the radio spectrum stays silent. There is nothing for a radio-based detector to find.
What the drone cannot hide is its motors. A multirotor propeller with B blades turning at R revolutions per second chops the air B times R times per second. Because the chopping is periodic but not sinusoidal, energy also appears at 2x, 3x, 4x that fundamental rate. The result is a harmonic comb: evenly spaced peaks in the frequency spectrum with gaps between them. Wind, traffic, and rain create broadband noise but do not produce an evenly spaced comb. VolAnti listens for that structure.
The Hardware: A 91 mm Box Built for Under 80 GBP
The README lists the main components: four ICS-43434 MEMS microphones in a plus-shaped array 79 mm corner to corner, an ESP32-S3 microcontroller, an e-paper display, a LoRa radio module, a beeper, an LED, and a haptic motor. The complete parts list for a full unit costs between 50 GBP and 80 GBP.
The microphone array uses four channels clocked from one source. Their streams are summed. At the frequencies of interest, 79 mm is about a tenth of a wavelength, so sound from any direction adds in phase and each capsule's own noise does not. The README states this gives roughly 6 dB of signal-to-noise improvement and is nearly omnidirectional. Beamforming was evaluated and found to provide no benefit at this array size, so the design uses sensitivity rather than directionality.
The e-paper display records the alert and the time even when power is cut, which matters for a device deployed in the field. Between alerts, the LED blinks slow green so a quiet unit and a dead unit never look the same. A snooze button silences the audio and haptic outputs for ten seconds while detection continues running underneath.
The Detection Algorithm: Four Independent Detectors on One Spectrum
The firmware runs a 2048-point FFT every 512 samples at 16 kHz, producing one spectrum frame every 32 ms. An adaptive noise floor tracks what the site normally sounds like at each frequency and subtracts it. Quiet backgrounds are learned in about 6 seconds; a passing vehicle or broadband burst updates the model in under a second.
Four independent detection tiers run on each spectrum frame:
Tier 1, called Fast comb, scores comb shapes against the fast noise floor. It catches aircraft that are arriving, approaching, or changing speed. The README states its latency is 0.23 seconds from first sound to alarm, and that this tier is sealed: golden test vectors pin its behaviour so the same audio produces the same score to the last decimal place on a laptop and on the hardware.
Tier 2, called Slow comb, uses the same scoring against a 30-second floor and targets aircraft that hover. Its latency is 1.4 to 4 seconds.
Tier 3, called Envelope wash, detects broadband modulation above 3 kHz caused by high-thrust, close flight. Latency is 1 to 3 seconds.
Tier 4, called No-floor comb, applies a 2-second Welch spectrum with whitening and no subtracted floor. It targets aircraft hovering in a location that never goes quiet, such as a persistent high-altitude drone. The README states the 104 m detection was achieved with this tier, with a latency of 5 to 15 seconds.
The track-and-hold mechanism requires the same candidate rate to win six consecutive frames within 2 percent variation before an alarm fires. Noise does not produce that pattern.
Alert Outputs and LoRa Networking
When any detection tier triggers, five outputs fire simultaneously. The beeper sounds a loud patterned alarm that the README describes as audible through a wall. The LED flashes red through a light pipe in the enclosure lid. A haptic motor pulses for whoever is holding the device. The e-paper display writes the alert and the time, and the display retains the information without power. A LoRa radio transmits an 18-byte packet to every other VolAnti unit in range.
The LoRa networking lets multiple units covering different positions share detection events. The README describes running several units and documents the deployment considerations in `docs/deploying.md`. A single battery charge runs the device for 18 to 22 hours according to the README.
The detection range the README documents is 104 m measured against a hovering airframe in light wind. That figure applies to the specific test conditions described in `docs/test-results.md`.
Building VolAnti: Two Paths
The repository provides two build guides. `docs/breadboard-build.md` is for evaluating the firmware on a breadboard before committing to a PCB. `docs/pcb-build.md` covers the full unit, with PCB files, a bill of materials, and component placement in `hardware/pcb/`. A printable enclosure is in `hardware/enclosure/`.
The firmware lives in `firmware/`. The reference detector implementation is in `src/`. Test audio files for verifying the algorithm without hardware are in `test/`. The website source is `index.html` and is deployed to `volantitech.com`, which also hosts a live simulator that lets you test the detection algorithm on a sound field you control in a browser.
The repository includes a `CITATION.cff` file for academic citation and an `environment.yml` for setting up the Python environment used for offline signal processing and analysis.
Limitations and Scope
VolAnti is an acoustic tripwire for the last few hundred metres around a fixed position. It is not radar and does not provide the range, bearing, or tracking capability that radar-based systems deliver. A quiet electric drone flying at altitude may stay below the detection threshold entirely, depending on the noise floor of the deployment site.
The README explicitly states: detection and alert only, no jamming, no interception, no countermeasures, ever. This scope constraint is intentional and documented.
The detection model assumes a multirotor propeller signature. Fixed-wing aircraft or exotic propulsion configurations that do not produce the expected harmonic comb pattern may not trigger detection.
A commercial comparison is DEDRONE, which is a radio-frequency and camera-based detection platform used in enterprise and government contexts. DEDRONE relies on radio fingerprinting, which is the approach that fails against fibre-optic FPV aircraft. VolAnti's acoustic approach covers the gap those systems leave but at much shorter range and without the tracking features a full-security deployment requires.
The no-license file in the repository's `LICENSES/` folder should be reviewed before deployment in a commercial context. The main `LICENSE.md` at the root governs the project.
Maintenance and Repository Layout
The last push to the repository was on 2026-09-20. The repository is not archived. The project includes a `CITATION.cff` for academic use, which suggests the author intends it to be used and cited in research contexts. The README includes a section titled "A note from Agam" and a contributing guide.
The repository layout puts firmware in `firmware/`, the reference Python detector in `src/`, test audio in `test/`, hardware files in `hardware/`, and documentation in `docs/`. The website source and live simulator are served from `index.html` and published at `volantitech.com`.
Editorial conclusion
VolAnti fits schools, farms, and small security operations that need a low-cost local tripwire for drones in the last few hundred metres, especially against fibre-optic FPV aircraft that emit no radio signal. It is not a replacement for radar or a full site security system. Before building, confirm that your environment is quiet enough: the detection model requires a consistent acoustic background and an adaptive noise floor that takes roughly six seconds to learn a quiet site and under a second for broadband bursts. The build comes in two forms: a breadboard version for evaluation and a full PCB unit documented in `docs/pcb-build.md`.
Frequently asked questions
Does VolAnti detect drones that use no radio?
Yes. VolAnti detects drones acoustically by listening for the harmonic comb that propellers produce, so it works against fibre-optic FPV aircraft that transmit no radio signal. The README states this is the design motivation: radio-based detectors fail against silent aircraft.
How far can VolAnti detect a drone?
The README documents a measured detection distance of 104 m against a hovering airframe in light wind. That result was from the Tier 4 no-floor comb detector. Detection range varies by flight mode, rotor size, and ambient noise at the deployment site.
What does it cost to build a VolAnti unit?
The README states that parts for a full unit cost between 50 GBP and 80 GBP. The PCB files, bill of materials, component placement, and a printable enclosure are all in the repository under `hardware/`.
Official sources
Add this badge to your README
If you maintain this project, the badge below links readers to this analysis and shows its maintenance status from the daily GitHub snapshot. Paste the markdown into your README; add ?metric=license or ?metric=stars to the image URL for a different field.
[](https://hysenlabs.com/projects/agamrossen-volanti)