🛡️ NetFenceAI: Intelligent Wi-Fi Threat Detection for Mobile

NetFenceAI is a functional, full-stack Proof-of-Concept (PoC) designed to detect, log, and visualize malicious wireless networks. It pairs a Flutter-based Android frontend with a local Python Flask backend to ingest real-time radio-frequency metrics and evaluate spatial threat signatures.

---

⚙️ Architecture & Data Flow

[ Flutter App ] ───(HTTP JSON: SSID, BSSID, RSSI, GPS)───> [ Python Flask Backend ]
       │                                                              │
[ UI Map Updates ] <───(Threat Score Vector [0-10])───────────────────[ Processing Engine ]

---

🛰️ Core Features

• Real-time Metadata Extraction: Scans and extracts raw hardware-level Wi-Fi parameters (SSID, BSSID/MAC, RSSI dBm values, encryption flags) and GPS coordinate streams.
• Dual-Layer Threat Scoring: * Primary: Deterministic rule-based engine (detects open networks, weak WEP encryption, MAC spoofing via 02: prefixes, and Evil Twins via SSID/MAC correlation).
  • Secondary: Optional Isolation Forest ML anomaly detection for behavioral pattern scoring.
• Spatial Threat Mapping: Renders visual danger vectors and coordinate intersections directly onto an interactive OpenStreetMap UI.
• Persistent Background Protection: Leverages WorkManager for continuous background scanning and pushes OS-level notifications for high-risk networks.
• Local Processing: Designed as an edge-computing PoC to keep spatial and network data localized and private.

---

📁 Repository Structure

• frontend/ - Flutter application source
  • android/ - Android Gradle configuration and native integration
  • lib/ - Dart app logic, UI screens, services, and models
• backend/ - Python Flask API and machine learning inference
  • app.py - Backend entry point
  • routes/ - API endpoints for scan ingestion and threat queries
  • models/ - Database and ML model helpers
  • test_data_generator.py - Local test dataset generator

---

🚀 Android App Build

Prerequisites

• Flutter SDK installed and configured
• Android SDK installed with API level 36
• Java JDK 11 or newer

Build Debug APK

cd frontend
flutter clean
flutter pub get
flutter build apk --debug

The generated APK is available at:

frontend\build\app\outputs\flutter-apk\app-debug.apk

Backend Setup

Prerequisites

• Python 3.8+
• Virtual environment support

Run Backend

cd backend
python -m venv venv
.\venv\Scripts\Activate.ps1
pip install -r requirements.txt
python app.py

The backend server exposes the API on http://127.0.0.1:5000 by default.

📝 Engineering Notes & Limitations

Geofencing: Active perimeter-crossing background geofence alerts are currently disabled due to library deprecation. Threat zones are instead visually rendered via coordinate mapping on the dashboard.

Android SDK: Current Android compile SDK is set to API 36 to ensure native plugin compatibility (wifi_scan, geolocator).

Environment: This is an edge-computing prototype designed to run locally. For enterprise deployment, the backend would require containerization (Docker) and cloud-hosted ML training pipelines.

Commit & Push

This repo is connected to:

https://github.com/csmdg7/NetFenceAI

Contact

For development or deployment questions, review the frontend and backend folders for implementation details.