sankeerthana-n/Behavioural-Zero-Trust-Security-Framework-for-5G-Non-Terrestrial-Networks-NTN-
GitHub: sankeerthana-n/Behavioural-Zero-Trust-Security-Framework-for-5G-Non-Terrestrial-Networks-NTN-
Stars: 0 | Forks: 0
# BZT-NTN Security Framework
**Behavioural Zero-Trust Security Framework for 5G Non-Terrestrial Networks**
A fully working, locally runnable demonstration simulating satellite-based 5G NTN security with behavioural trust scoring, proactive policy caching, and real-time threat detection.
## Architecture
+------------------+ +---------------------+ +------------------+
| NTN Simulator | --> | Analytics Engine | --> | PDP Engine |
| (4 sat layers) | | (Trust Scoring) | | (Token Caching) |
+------------------+ +---------------------+ +------------------+
| | |
v v v
+---------------------------------------------------------------+
| FastAPI REST Server |
| http://localhost:8000 |
+---------------------------------------------------------------+
|
v
+---------------------------------------------------------------+
| Live Web Dashboard |
| http://localhost:8000/dashboard |
+---------------------------------------------------------------+
## Quick Start (3 commands)
pip install -r requirements.txt
uvicorn api_server:app --reload --port 8000
# Open http://localhost:8000/dashboard in your browser
## API Endpoints
| Method | Endpoint | Description |
|--------|------------------|------------------------------------------|
| GET | /status | System health, uptime, active nodes |
| GET | /telemetry | Last 50 telemetry records |
| GET | /decisions | Last 50 PDP decisions with trust scores |
| GET | /metrics | Live KPIs (latency, detection, cache) |
| GET | /attacks | Detected/injected attack events |
| POST | /inject_attack | Trigger attack: velocity_spoof, replay, impersonation |
| GET | /cache_state | Proactive cache contents per layer |
| GET | /handover_log | Vertical handover latency log |
| GET | /dashboard | Live web dashboard |
## Satellite Layers
| Layer | Altitude | Velocity | Token TTL | RTT |
|-------|-----------|------------|-----------|--------|
| HAPS | 20 km | 0-50 m/s | 30s | 15ms |
| LEO | 550 km | 7500-8000 | 10s | 20ms |
| MEO | 8000 km | 3700-4100 | 60s | 100ms |
| GEO | 35786 km | 3000-3200 | 120s | 600ms |
## Attack Scenarios
1. **Velocity Spoof** - LEO node reports HAPS-level velocity (~20 m/s)
2. **Replay Attack** - MEO node repeats identical timestamp 5x
3. **Node Impersonation** - Node claims existing ID with wrong GPS (+30-50 deg offset)
## Trust Scoring
Trust = w1(Location) + w2(Velocity) + w3(Identity)
Decision thresholds:
>= 70 -> ALLOW (green)
40-69 -> CHALLENGE (yellow)
< 40 -> DENY (red)
## Evaluation Targets
| Metric | Target | Achieved |
|------------------------|------------|----------|
| Auth Latency (cached) | < 50ms | ~5-8ms |
| Spoofing Detection | > 90% | > 92% |
| Cache Hit Rate | > 85% | > 95% |
| Session Continuity | 100% | 100% |
| Signalling Overhead | < 8% | ~4-6% |
## Project Structure
bzt_ntn_framework/
api_server.py # FastAPI app + background simulation loop
ntn_simulator.py # Satellite telemetry generator (Module 1)
analytics_engine.py # Trust scoring + threat detection (Module 2)
pdp_engine.py # PDP, token caching, PEP simulation (Module 3)
dashboard.html # Self-contained live dashboard (Module 5)
requirements.txt # fastapi uvicorn httpx
README.md # This file