AI Network System

AI-driven network intelligence for high-density environments.

Project Structure

• backend/ - Team 2 FastAPI API and slicing decision engine
• model/ - Team 1 labeling, training, and prediction modules
• simulation/ - Team 1 simulation module
• docs/ - project handoff notes and quick guides

Problem

In places like colleges, stadiums, and public events, many users connect at once. This causes latency, unstable connectivity, and poor service for critical communication.

Solution

This project combines:

• Simulation of network conditions
• AI-based prediction using Team 1 model
• Intelligent bandwidth allocation using 5G slicing logic

Team 2 Backend (Krishna)

The backend service adds these APIs:

• POST /simulate
• POST /predict
• POST /allocate
• GET /health

Setup

python -m venv .venv
.venv\Scripts\activate
pip install -r requirements.txt

Run

uvicorn backend.app:app --reload --host 0.0.0.0 --port 8000

Open docs at:

• http://127.0.0.1:8000/docs

Detailed handoff document:

• docs/read-this-first.md

Model Integration

Team 1 output model should be available at:

• model/model.pkl

Optional custom path:

set MODEL_PATH=D:\path\to\model.pkl

If the model is unavailable or fails at runtime, prediction uses heuristic fallback so demo still works.

API Quick Examples

Simulate

curl -X POST "http://127.0.0.1:8000/simulate" -H "Content-Type: application/json" -d "{}"

Predict

curl -X POST "http://127.0.0.1:8000/predict" -H "Content-Type: application/json" -d "{\"zone\":\"Stadium\",\"signal_strength\":56,\"bandwidth_usage\":7.2,\"latency\":180,\"packet_loss\":3.1,\"num_users_in_zone\":250,\"time_of_day\":20}"

Allocate (manual label)

curl -X POST "http://127.0.0.1:8000/allocate" -H "Content-Type: application/json" -d "{\"total_bandwidth_mbps\":1000,\"congestion_label\":\"HIGH\",\"demand_mbps\":{\"high_priority_mbps\":600,\"medium_priority_mbps\":450,\"low_priority_mbps\":400}}"

Allocate (prediction-assisted)

curl -X POST "http://127.0.0.1:8000/allocate" -H "Content-Type: application/json" -d "{\"total_bandwidth_mbps\":1000,\"demand_mbps\":{\"high_priority_mbps\":520,\"medium_priority_mbps\":340,\"low_priority_mbps\":280},\"prediction_features\":{\"zone\":\"Academic_Block\",\"signal_strength\":60,\"bandwidth_usage\":6.5,\"latency\":140,\"packet_loss\":2.2,\"num_users_in_zone\":210,\"time_of_day\":18}}"

Reality Statement

This system simulates network behavior and provides AI decision support. It does not directly control real telecom infrastructure.

Frontend Interface (Team 3)

A modern, real-time React dashboard has been integrated to visualize the network traffic, run AI predictions, and execute 5G slicing logic interactively.

Network Monitoring & Dashboard

AI Congestion Prediction

5G Slicing Allocator