😴 Sleep Quality Estimation Using Wearable Sensors

A Cyber-Physical System (CPS) and Machine Learning project designed to provide real-time, privacy-focused sleep quality estimation. This system integrates hardware sensing with a Random Forest model to process physiological data locally (Fog Computing), delivering an interpretable sleep score from 0-100.

---

🚀 Project Overview

Traditional sleep monitoring often relies on cloud-based processing, leading to latency issues and privacy concerns. This project addresses these gaps by:

Feature	Description
🔒 Local Processing (Fog Node)	Uses a laptop as a processing hub to ensure data privacy and low-latency feedback
⚡ Real-Time Sensing	Captures raw PPG (Photoplethysmogram) data via an Arduino-integrated pulse sensor
🧠 Interpretable AI	Implements a Random Forest Classifier to categorize sleep quality and identify key physiological drivers

---

🏗️ System Architecture

The project is structured into three distinct layers:

┌─────────────────────────────────────────────────────────────────────┐
│                        CYBER/FOG LAYER                              │
│  ┌─────────────┐   ┌─────────────┐   ┌─────────────────────────┐   │
│  │ Preprocessing│ → │  ML Model   │ → │  Streamlit Dashboard    │   │
│  │ (HeartPy)   │   │(RandomForest)│   │  (Real-time Visuals)   │   │
│  └─────────────┘   └─────────────┘   └─────────────────────────┘   │
└─────────────────────────────────────────────────────────────────────┘
                              ↑
                    USB Serial (PySerial)
                              ↑
┌─────────────────────────────────────────────────────────────────────┐
│                      PHYSICAL LAYER                                 │
│           ┌──────────────────────────────────────┐                 │
│           │  Arduino UNO + PPG Pulse Sensor      │                 │
│           │  (Raw Physiological Data Acquisition)│                 │
│           └──────────────────────────────────────┘                 │
└─────────────────────────────────────────────────────────────────────┘

Layer Details:

1. Physical Layer: PPG Pulse Sensor + Arduino UNO for raw physiological data acquisition
2. Communication Layer: USB Serial bridge using PySerial for high-speed data transmission
3. Cyber/Fog Layer:
   • Preprocessing: Feature extraction (BPM, HRV) from raw signals
   • Inference: A pre-trained Random Forest model for classification and scoring
   • Visualization: A Streamlit-based real-time dashboard

---

🛠️ Tech Stack

Category	Technologies
Hardware	Arduino UNO R3, PPG Pulse Sensor, LEDs, Resistors
Languages	Python 3.9+, C++ (Arduino Sketch)
ML Framework	Scikit-learn (Random Forest)
Visualization	Streamlit
Key Libraries	pandas, joblib, pyserial, heartpy, numpy, matplotlib

---

📂 Project Structure

Sleep_Quality_Project/
├── 📁 data/                    # Training datasets
│   └── Sleep_health_and_lifestyle_dataset.csv
├── 📁 firmware/                # Arduino source code (.ino)
├── 📁 models/                  # Saved Machine Learning models
│   └── sleep_quality_model.pkl
├── 📁 scripts/                 # Python scripts for ML and Dashboard
│   ├── train.py               # Model training and feature importance
│   └── dashboard.py           # Streamlit real-time interface
├── 📄 requirements.txt        # Python dependencies
├── 📄 .gitignore              # Git ignore rules
├── 📄 LICENSE                 # MIT License
└── 📄 README.md               # Project documentation (you are here!)

---

⚙️ Installation & Setup

1. Hardware Connection

1. Connect the PPG Pulse Sensor to the Arduino:

   • Signal → A0
   • VCC → 5V
   • GND → GND

2. Connect the Arduino to your laptop via USB

2. Software Installation

Clone the repository:

git clone https://github.com/GuruMohith24/Sleep_Quality_Project.git
cd Sleep_Quality_Project

Set up Virtual Environment:

python -m venv .venv

# Activate on Windows:
.venv\Scripts\activate

# Activate on Mac/Linux:
source .venv/bin/activate

Install Dependencies:

pip install -r requirements.txt

---

📈 Usage

1. Upload Firmware

Upload the firmware/ppg_sensor.ino to your Arduino using the Arduino IDE.

2. Train Model (Optional)

Run the training script to see feature importance and save the model:

python scripts/train.py

Expected Output:

Loaded data from: .../data/Sleep_health_and_lifestyle_dataset.csv

--- Feature Importance ---
Sleep Duration              0.35
Heart Rate                  0.30
Physical Activity Level     0.20
Daily Steps                 0.10
Age                         0.05

Model saved to: .../models/sleep_quality_model.pkl

3. Run Dashboard

Launch the real-time Streamlit interface:

streamlit run scripts/dashboard.py

---

🧠 Machine Learning Insights

The model uses a Random Forest Classifier which provides high robustness against noisy sensor data.

Feature Importance

Feature	Weight	Description
🛏️ Sleep Duration	~35%	Primary predictor of sleep quality
❤️ Heart Rate	~30%	Key physiological indicator
🏃 Physical Activity	~20%	Daily exercise impact
👣 Daily Steps	~10%	Movement patterns
📅 Age	~5%	Demographic factor

Model Output

• Classification: Binary (Good/Poor Sleep Quality)
• Score: Heuristic score from 0-100
• Threshold: Sleep quality ≥ 7 classified as "Good"

---

🔬 Why This Approach?

Challenge	Our Solution
Privacy Concerns	All processing happens locally on the Fog Node
Latency Issues	Edge computing eliminates cloud round-trips
Interpretability	Random Forest provides feature importance insights
Cost	Low-cost Arduino + open-source software stack

---

🎯 Future Enhancements

• Add LSTM/RNN for temporal pattern recognition
• Integrate SpO2 sensor for oxygen saturation monitoring
• Mobile app for remote monitoring
• Sleep stage classification (REM, Deep, Light)

---

👨‍💻 Author

Guru Mohith

• GitHub: @GuruMohith24

---

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

---

Made with ❤️ for better sleep quality monitoring