| Hi HN, I built SentryRF because I was frustrated by the lack of granular control on Android for identifying unwanted tracking devices. While Google's native "Unknown Tracker Alerts" are a good start, they often feel like a black box with limited diagnostic tools. The Tech: Signal Analysis: Instead of just listing every BLE/wifi device, SentryRF uses signal strength (RSSI) and temporal patterns to distinguish between a "passing" device (someone walking by with an AirTag) and a "following" device (a tracker on your vehicle), even peeping-toms. Hardware Sensors: I’ve integrated the magnetometer and ambient light sensors to help find non-broadcasting devices (like "dead" GPS units or wired pinhole cameras) by detecting magnetic anomalies and infrared emissions. Sound Locator: I implemented an audio-guided proximity algorithm that increases beep frequency as you approach the source. It’s essentially a Geiger counter for Bluetooth and wifi signals. Privacy (The most important part): Zero Cloud: There is no backend. No scan data, location data, or telemetry ever leaves the device. No Accounts: You don’t need to sign up. I don't want your email or your name. Permissions: It requires quiet a few permissions, but I’ve documented exactly why each is needed in the app’s onboarding. Why I’m showing it here:
I’m looking for technical feedback on the signal smoothing I’m using for the proximity tracker and any edge cases I might have missed regarding the rogue cell tower (IMSI catcher) detection. The app has a 7-day free trial so you can test the "Pro" features (like the Sound Locator) without paying a cent. I'll be around all day to answer questions about the stack or the detection logic! |
Core Capabilities: Multi-Protocol Detection: Scans the 2.4GHz spectrum to identify AirTags, Tiles, SmartTags, and unbranded BLE trackers using high-fidelity manufacturer data parsing.
Behavioral AI Engine: Employs a local neural network to differentiate between "ambient noise" (a neighbor's headphones) and "threat patterns" (a device following your trajectory across multiple GPS coordinates).
Tactical Sound Locator: Uses an RSSI-weighted triangulation engine and haptic/audio feedback to guide users to a hidden device's physical location through signal gradient analysis.
Deep Hardware Analysis: Identifies rogue WiFi "Evil Twins," IMSI catchers, and unauthorized IoT devices using OUI lookups and chipset-family fingerprinting.
Privacy-Centric Architecture: All analysis, including ML retraining and database storage, happens 100% on-device. No RF data or location history ever leaves the handset.