DroneDetectAndroid

Android-side utilities for rotor detection and Wi‑Fi reconnaissance. Assets ship as base64 to keep the repo lean for air-gapped deployments and are restored automatically during Gradle builds.

Bootstrap

After cloning, restore the binary assets (TensorFlow Lite model and PNG drawables):

python3 scripts/bootstrap_assets.py

Gradle also depends on this script via the preBuild hook, so IDE and CI builds will automatically decode assets into app/src/main/... if Python is available.

Pipeline

1. Asset bootstrap – Decode the model and icons from assets/*.b64 using the script above (or let Gradle call it).
2. Runtime wiring – MainActivity requests Wi‑Fi/location permissions, wires a simple UI, and calls DroneSignalDetector.startScan().
3. Scan handling – DroneSignalDetector schedules repeated WifiManager.startScan() calls, ingests stub flight data, and surfaces scan results plus the last telemetry snapshot to the UI via callbacks.

Building

This repo ships a minimal Gradle wrapper script for offline/air-gapped work. Supply a compatible gradle/wrapper/gradle-wrapper.jar (from Gradle 8.2.1) in your environment, then run:

./gradlew assembleDebug

If you prefer a hosted toolchain, open the project in Android Studio (Giraffe+), accept SDK prompts, and build the app module. Remember to run the asset bootstrap script if Python is missing from your build host.

Testing

To validate the asset bootstrap step:

python3 scripts/bootstrap_assets.py

Each asset should report a decoded message. Remove the generated app/src/main/assets and app/src/main/res/drawable contents after testing to keep the repo text‑only.