🔐 Side-Channel Attack Detection using Machine Learning

Overview

This project detects side-channel attacks in IoT devices using Machine Learning. It processes input data, applies scaling and PCA, and predicts whether an attack is present or not.

---

Live Demo

👉 Try the app here: 🔗 https://huggingface.co/spaces/Tejaswa18/Secure-IoT-ML-Model

---

Project Files

• Main (2).ipynb → Model training and preprocessing
• app.py → Streamlit web application
• model.pkl → Trained model
• scaler.pkl → Data scaler
• pca.pkl → PCA transformation
• test_sample.csv → Sample input
• requirements.txt → Dependencies

---

Technologies Used

• Python
• Scikit-learn
• Pandas, NumPy
• Streamlit

---

How It Works

1. Input data is taken from user or CSV
2. Data is scaled using scaler.pkl
3. PCA is applied using pca.pkl
4. Model (model.pkl) predicts the result
5. Output shows whether attack is detected

📊 Testing

Use test_sample.csv or enter values manually in the app.

---

Features

Simple and fast prediction Lightweight model Easy to run locally

---