CellSeeU - Privacy Awareness Dashboard

in-progress

Comprehensive cellular, WiFi, and wireless tracking detection system with Android app, Flask backend, and real-time WiFi triangulation

Technologies

Android Java Python 3 Flask Leaflet.js OpenCelliD API GPS Compass Sensors

Tags

PrivacySecurityAndroidPythonFlaskWiFiMapping

What Does This Do?

CellSeeU is a privacy awareness platform that reveals which services can track your mobile device in real-time by detecting:

  • 📡 Cell towers that can see your IMEI number
  • 📶 WiFi networks that detect your MAC address and probe requests
  • 🧭 WiFi router locations estimated using directional triangulation
  • 🗺️ Detection radius visualization showing tracking coverage

The Problem

Most smartphone users don’t realize how many services track their devices continuously through:

  • Cell tower registration (IMEI tracking)
  • WiFi probe requests (MAC address broadcasting)
  • Network triangulation (position estimation)

CellSeeU makes the invisible visible by showing you exactly which towers and networks are tracking your movements in real-time.

Key Features

🔍 Privacy Tracking Detection

  • IMEI Detection Radius: Visual circle showing which cell towers can track you
  • WiFi MAC Tracking: Shows which networks detect your WiFi probe requests
  • Real-time Dashboard: Live updates as you move around
  • Tracking Panel: Lists all services currently detecting your device

🧭 WiFi Triangulation System

Advanced algorithm that estimates WiFi router positions using:

  1. Your GPS position (latitude, longitude)
  2. Signal strength (RSSI in dBm)
  3. Compass heading (direction you’re facing)
  4. Multiple scan points (walking triangulation)

Accuracy:

  • 2 scans: ±50-100m
  • 3-5 scans: ±20-50m
  • 10+ scans: ±10-30m

📱 Android App Integration

Custom scanner app that collects:

  • Cell tower data (IMEI, signal strength, carrier, network type)
  • WiFi networks (SSID, BSSID, signal, security)
  • Orientation sensors (compass heading, pitch, roll)
  • GPS location tracking
  • Uploads combined data to Flask backend

Technical Implementation

System Architecture

Android App (Scanner)        Backend (Flask)         Frontend (Dashboard)
├─ CellTowerScanner.java  →  /api/towers/upload  →  Dashboard Tab
├─ WiFiScanner.java       →  /api/wifi           →  WiFi Networks Tab
├─ OrientationSensor.java →  /api/wifi/positions →  Triangulated Map
└─ GPS Location              (Processing)           (Visualization)

Backend Services (Python/Flask)

Core APIs:

  • /api/towers - Get detected cell towers
  • /api/towers/upload - Upload scan data from Android
  • /api/towers/nearby - Get OpenCelliD towers in viewport
  • /api/wifi - Get WiFi networks from last scan
  • /api/wifi/positions - Get triangulated WiFi AP positions

Processing Services:

  • tower_location.py - OpenCelliD integration for tower geolocation
  • carrier_lookup.py - MCC/MNC to carrier name mapping
  • wifi_triangulation.py - Directional positioning algorithm

WiFi Triangulation Algorithm

Physics: WiFi signals are directional - if you know your position, direction facing, and signal strength, you can estimate router location.

Process:

Scan #1: Position (53.2919, -6.6860), facing NE (45°), signal -42 dBm
         → Router ~25m away in NE direction
         → Estimated position: (53.2921, -6.6858)

Scan #2: Position (53.2925, -6.6865), facing E (90°), signal -38 dBm
         → Router ~20m away in E direction
         → Estimated position: (53.2925, -6.6863)

Scan #3: Position (53.2930, -6.6870), facing SE (135°), signal -45 dBm
         → Router ~30m away in SE direction
         → Estimated position: (53.2928, -6.6868)

Triangulation: Average all positions → (53.2925, -6.6863) ±15m

Factors Affecting Accuracy:

  • Compass calibration
  • WiFi reflections (multipath)
  • Walls and obstacles
  • Device orientation changes

Frontend (Interactive Dashboard)

Technologies: HTML, JavaScript, Leaflet.js (maps)

Pages:

  1. Dashboard - Statistics, tower list, tracking panel
  2. Map - Interactive map with:
    • Cell tower markers
    • IMEI detection radius circles
    • WiFi AP position markers (color-coded by confidence)
    • Device location tracking
  3. WiFi - Network list with security info, signal strength, estimated positions

Features:

  • Real-time auto-refresh (configurable interval)
  • Confidence indicators (high/medium/low)
  • OpenCelliD integration for tower verification
  • Responsive design for mobile/desktop

Security & Privacy Features

Built-in Security:

  • 🔒 Flask-Talisman: HTTPS enforcement, security headers
  • 🛡️ CORS Configuration: Restricted API origins
  • 🔐 Input Validation: All API endpoints validate data
  • 🚨 Rate Limiting: Prevent API abuse
  • 📝 Environment Variables: Secure configuration management
  • ✅ Security Testing: Bandit + Safety in test suite

Privacy by Design:

  • No data stored permanently (in-memory only)
  • 100-scan history limit (rolling buffer)
  • No user accounts or tracking
  • Local network deployment (no cloud)
  • Educational purpose only

Technology Stack

Android App:

  • Language: Java
  • Components: CellTowerScanner, WiFiScanner, OrientationSensor, ApiClient
  • Build: Gradle, Android SDK
  • Sensors: GPS, Compass, Accelerometer, TelephonyManager, WiFiManager

Backend:

  • Framework: Flask 3.0.0
  • Libraries:
    • flask-cors (API access)
    • flask-talisman (security headers)
    • requests (OpenCelliD API)
    • python-dateutil (timestamp handling)
  • External APIs: OpenCelliD (cell tower geolocation)

Frontend:

  • Mapping: Leaflet.js
  • UI: Bootstrap 5
  • Data: Real-time AJAX updates
  • Visualization: OpenStreetMap tiles

Testing & Quality:

  • Testing: pytest, pytest-flask, pytest-cov, pytest-mock
  • Security: bandit (static analysis), safety (dependency scanning)
  • Code Coverage: pytest-cov with .coverage tracking

Use Cases

1. Privacy Education

Demonstrate to users how cell towers and WiFi networks track devices continuously, even when not connected.

2. Security Research

Analyze tracking capabilities of cellular and WiFi infrastructure for security audits.

3. Network Analysis

Map WiFi access point locations using crowdsourced triangulation data.

4. Geolocation Research

Study accuracy of WiFi-based positioning systems compared to GPS.

Development Workflow

Setup Backend:

cd path/to/cell_see_u
.\.venv\Scripts\Activate.ps1
python app.py

Build Android App:

cd android_scanner
.\gradlew assembleDebug
# APK: android_scanner\app\build\outputs\apk\debug\app-debug.apk

Configure:

  • Backend: Edit .env (SERVER_IP, OPENCELLID_API_KEY)
  • Android: Edit ServerConfig.java (SERVER_URL)

Current Status & Future Enhancements

✅ Implemented:

  • Cell tower scanning and detection radius
  • WiFi network scanning with security info
  • Orientation sensor integration (compass, pitch, roll)
  • WiFi triangulation service with confidence scoring
  • Interactive map with multiple layers
  • Real-time dashboard with auto-refresh
  • OpenCelliD integration for tower verification
  • Security headers and CORS configuration

🚧 In Progress:

  • Improve triangulation accuracy with Kalman filtering
  • Add historical tracking visualization (movement path)
  • Export data to CSV/JSON for analysis

💡 Future Ideas:

  • Bluetooth beacon detection
  • Cell tower handover tracking
  • Signal strength heatmaps
  • Multi-device comparison mode
  • Privacy score calculation

Technical Highlights

Algorithm Innovation:

Custom WiFi triangulation using compass heading + GPS + RSSI - unique approach combining orientation sensors with traditional positioning.

Full-Stack Integration:

Seamless data flow from Android sensors → Flask API → Interactive web dashboard with real-time updates.

Privacy-First Design:

Educational tool that raises awareness without storing sensitive data permanently.

Cross-Platform:

Android app + Python backend + Web frontend = comprehensive privacy tracking solution.

Educational Value

This project demonstrates:

  • Mobile Development: Android app with sensor integration
  • Backend Development: Flask API with external service integration
  • Algorithm Implementation: WiFi triangulation mathematics
  • Security Best Practices: HTTPS, CORS, input validation, security headers
  • Privacy Awareness: Understanding device tracking mechanisms
  • Testing: Comprehensive test suite with security analysis
  • Documentation: Clear README, API docs, setup guides

Notes

  • Local Development: Runs on private network (e.g., http://192.168.x.x:5000)
  • Educational Purpose: Designed to raise privacy awareness, not for malicious tracking
  • No GitHub Repository: Private project not yet published
  • Android App: Requires manual APK installation (not on Play Store)
  • OpenCelliD API: Free tier has rate limits (requires API key)
← Back to Projects