Skip to content
plusmaliasPublic

About

Android counter-surveillance app: detects AirTags & item trackers, followers, hidden cameras, stalkerware, skimmers, IMSI-catchers and police/emergency signals using only your phone's radios. Offline, encrypted, disguisable.

Topics

Resources

Stars

5 stars

Watchers

1 watching

Forks

Repository files navigation

📡 RFWatch

See the radio world around you. Know if something is watching you.

An Android counter-surveillance and RF situational-awareness app. RFWatch passively listens to the Bluetooth, Wi-Fi and cellular signals around you, works out roughly where they're coming from, flags item trackers, hidden cameras, stalkerware and surveillance gear, and warns you when something seems to be following you.

Platform Kotlin Version Offline Encrypted License: GPL v3

⬇️ Download the APK · Features · Install · Privacy · How it works · Build


Why RFWatch?

Cheap trackers, hidden cameras and phone spyware have made covert surveillance easy. For people in abusive relationships, being stalked, or simply wanting to know what's around them, "is someone tracking me?" can be a serious safety question.

RFWatch answers it with the radios already in your phone. There's no extra hardware, no account and no cloud. It's built to be honest about what a phone can and can't detect: every result carries a confidence level, and when the evidence is weak it says so rather than guessing.

Designed in Australia. Metric units, local magnetic declination, Australian emergency-service and fleet vendor rules, and Australian support services built in.


✨ Features

🎯 Trackers & followers

  • Item-tracker detection by payload, not MAC address, so it keeps working as trackers rotate their addresses:
    • Apple AirTag / Find My
    • Samsung SmartTag
    • Tile
    • Chipolo
    • Google Find My Device (FMDN)
  • Follower detection: flags a device that keeps turning up with you across different places and times.
  • Approach / recede trends: arrows show whether a signal is getting closer or moving away.
  • "May be the same device" hints: links Bluetooth address rotations and multi-radio routers so one device doesn't look like many.
  • Home watch: notices new devices that appear and persist around home.
  • GPS-tracker and drone (Remote ID) recognition.

🛡️ Protect hub: on-demand checks

Check What it does
Room sweep Guided sweep of a room for cameras, IoT gear and unknown transmitters
Car sweep Checks a vehicle for hidden trackers
Camera finder Two modes: an IR finder (front camera sees night-vision LEDs) and a lens finder (torch + back camera catch the glint of a hidden lens)
Stalkerware check Scans installed apps against public stalkerware/spyware indicators (package names, signing certificates, APK hashes) and flags apps holding risky permissions
Skimmer check Looks for Bluetooth signatures common to card skimmers
Ultrasonic beacon check Listens briefly (17–22 kHz) for inaudible tracking beacons
Cell / IMSI-catcher check Heuristic "stingray" detection from cell identity and behaviour
Location scope North-up plot of where nearby devices most likely are. Only devices with a real position estimate are drawn, never invented angles

🚓 Awareness alerts

  • Police, ambulance and fire recognition, including Axon body-cam / TASER signatures and fleet Wi-Fi. These are the alerts that come through loud and clear.
  • Evil-twin Wi-Fi (a fake copy of a network you trust) and common wireless attack-tool signatures.
  • Hidden cameras and IoT devices by vendor and broadcast fingerprint.

🧭 Find it

  • Radar: live, heading-up radar with scan-health readout, distance rings, device icons, trend arrows and a "nearby now" list.
  • Hunt mode: a hotter/colder tone guides you to a device.
  • Heading sweep: turn slowly on the spot and RFWatch estimates the direction.
  • Drive-by locate: GPS-tagged readings estimate a fixed transmitter's position, with an error ellipse.
  • Map and history: where each signal was heard over time (MapLibre + OpenStreetMap in the maps build).

🗂️ Organised and easy to use

  • Devices list with filters (time window, threats, new, named, hide mine), multi-kind selection, sorting and nicknames.
  • One-tap cross-links from any device to Details → Locate → Hunt → Map.
  • Quick Settings tile and home-screen widget.

🤫 Discreet mode: built for personal safety

  • Disguise: the app can appear as a working Calculator. Type your PIN into the calculator to open RFWatch.
  • Neutral notifications that don't say what the app is.
  • Shake to hide.
  • App lock with a PIN or biometrics.
  • Evidence export: an HTML report plus CSV/JSON, with sightings, alerts, a timeline and map links, ready to hand to police, a lawyer or a support worker.

📲 Install

  1. Go to Releases and download an APK:
    • RFWatch-0.3.0-maps.apk: recommended. Live street maps; uses the internet only for map tiles.
    • RFWatch-0.3.0-offline.apk: maximum privacy. The app has no internet permission at all.
  2. Open the downloaded file on your phone. If Android says it can't install unknown apps, tap Settings, allow "Install from this source", then go back and tap Install.
  3. Open RFWatch and grant Location, Nearby devices / Bluetooth and Notifications. It can't hear anything without them.
  4. Keep Bluetooth and Location switched on while scanning.

Requirements: Android 8.0 (API 26) or newer. Tested on Android 14.

These are early test builds (debug-signed), not Play Store releases. Expect rough edges, and please report what you find.


🔒 Privacy & security

  • Your data never leaves your phone. RFWatch has no accounts, no analytics and no cloud.
  • The offline build has no INTERNET permission, so it physically cannot send data anywhere.
  • The maps build only requests the OpenStreetMap tiles for the area on screen. Scan records stay on the device.
  • Encrypted at rest with SQLCipher. The key is wrapped by the Android Keystore (hardware-backed where available).
  • Automatic retention purge (default 30 days) and a wipe-all-data button.
  • Excluded from cloud backup and device transfer.

🔬 How it works, honestly

A phone has omnidirectional antennas and no direction-finding array, so it can't measure a true bearing in an instant. Every direction in RFWatch is an estimate, and every estimate carries a confidence value.

Method How What you get
Heading sweep (standing still) Turn 360° with the phone against your chest; your body shadows the signal A bearing from a cosine-lobe fit, with a confidence score
Drive-by (moving) GPS-tagged signal readings fit a path-loss model A position with an error ellipse, or "mobile, trending"
Wi-Fi RTT (supported access points) 802.11mc round-trip timing A real distance used as a hard constraint

If a fit is too weak, RFWatch says "no clear direction" instead of drawing a fake dot. Poor GPS fixes are ignored for direction, and gaps in scanning aren't mistaken for a device leaving.

Cell / IMSI-catcher detection is heuristic. A normal (non-rooted) phone can't read baseband signalling, so RFWatch reasons from cell identity, radio type, neighbour cells and signal strength:

  • a 2G downgrade
  • a lone cell
  • an unknown cell
  • a LAC/TAC mismatch
  • rapid reselection
  • implausibly strong signal

A HIGH result means worth checking, not proof.

Tracker identifiers

Classified by advertisement payload and checked against the Bluetooth SIG UUID registry, the IEEE OUI registry and published research:

  • Apple Find My: 0x004C type 0x12; accessory 0xFD44
  • Samsung SmartTag: 0xFD5A / 0xFD59
  • Tile: 0xFEED / 0xFEEC / 0xFD84
  • Chipolo: 0xFE33 / 0xFE65
  • Google FMDN: 0xFEAA frames 0x40/0x41

Undocumented fields are treated as best effort, and the code says so.

Architecture

  • :core: pure Kotlin/JVM with no Android dependencies. All the RF maths, parsers, scoring and detection rules live here and are covered by 130+ unit tests that run on a bare JDK.
  • :app: Android (Kotlin + Jetpack Compose). Scanning, sensors, the encrypted database, alerts and UI. It wraps :core and never re-implements the maths.

Detection rules ship as JSON rule packs (trackers, law_enforcement, emergency_services, fleet, cameras_iot, attack_tools), alongside vendor and stalkerware indicator databases.

Standing on the shoulders of

AirGuard (TU Darmstadt) for tracker and follower detection; SnoopSnitch, AIMSICD, CellGuard and Crocodile Hunter for IMSI-catcher research; WiGLE for wardrive mapping; and the Echap / MVT stalkerware indicator projects.


🛠️ Build from source

You'll need JDK 17 and the Android SDK (platform 34). See BUILD.md for the full guide.

./gradlew :core:test                 # detection logic tests (JDK only)
./gradlew assembleOfflineDebug       # no-internet build
./gradlew assembleMapsOnlineDebug    # build with live maps

🆘 If you're in danger

RFWatch can help you gather evidence, but it is not a substitute for help from people. If you think someone is tracking or monitoring you:

  • Export your evidence first.
  • Think carefully before removing a tracker or app, because the person may notice.
  • Talk to a specialist service.
Australia
Emergency 000
1800RESPECT (domestic, family & sexual violence) 1800 737 732
MensLine Australia 1300 78 99 78
DVConnect (QLD) 1800 811 811

⚖️ Responsible use

RFWatch is for protecting yourself, your own devices, your property and your network. It records the presence of nearby devices, so treat that data with care. Surveillance and privacy laws vary by state and country, and you are responsible for using it lawfully. Detection is probabilistic: no result is proof on its own, and the absence of an alert is not a guarantee that you aren't being tracked.


📜 License

Copyright © 2026 Jorden Armstrong.

RFWatch is free software, licensed under the GNU General Public License v3.0. You can use it, study it, change it and share it. If you distribute a modified version, you must release your source code under the same licence and keep this credit.

For commercial licensing or partnerships, please get in touch via GitHub.


Built by Jorden Armstrong. Feedback, bug reports and field results are very welcome: open an issue.

About

Android counter-surveillance app: detects AirTags & item trackers, followers, hidden cameras, stalkerware, skimmers, IMSI-catchers and police/emergency signals using only your phone's radios. Offline, encrypted, disguisable.

Topics

Resources

Stars

5 stars

Watchers

1 watching

Forks

Releases

Packages

Contributors

Languages