OCF-IoTivity Smart Home Energy & Comfort Optimization System

An Open Connectivity Foundation (OCF) based prediction-assisted optimal control framework for smart homes that optimizes energy consumption while maximizing occupant thermal comfort. This is the first scalable, secure, and interoperable optimal control solution for smart-home IoT networks.

🌟 Key Achievements

36.82% Energy Savings - Significant reduction in HVAC energy consumption
3.36ms Round Trip Time - Ultra-low latency for real-time control
5s Average Response Time - Fast actuator control and adjustment
R² > 0.94 - High accuracy in energy and environmental parameter prediction

🎯 Overview

Traditional Home Energy Management Systems (HEMS) face significant challenges:

Interoperability issues across heterogeneous IoT devices
Energy vs. Comfort trade-off not adequately addressed
Scalability limitations with proprietary protocols
Security and privacy concerns in smart home networks

Our solution addresses these challenges through:

OCF-IoTivity Framework - Standardized, secure device connectivity
Edge Intelligence - LSTM models deployed on IoT devices for real-time predictions
Improved Firefly Algorithm - Enhanced optimization with inertia weight approach
Dual-Objective Optimization - Balances energy efficiency and occupant comfort

✨ Features

Core Capabilities

✅ Real-time Environmental Monitoring
- Temperature, humidity, air velocity sensing
- Mean radiant temperature measurement
- Energy consumption tracking
✅ Predictive Analytics
- LSTM-based energy consumption prediction
- Environmental parameter forecasting
- PMV (Predicted Mean Vote) comfort prediction
✅ Intelligent Optimization
- Modified Firefly Algorithm with logarithmic inertia weight
- Multi-objective optimization (energy + comfort)
- User preference-based control (α parameter)
✅ OCF Connectivity
- Seamless peer-to-peer device communication
- Protocol-agnostic interoperability
- Secure device on-boarding and provisioning
✅ Mobile Application
- Real-time environment monitoring
- User preference configuration
- Energy consumption visualization

🏗️ System Architecture

Active Chilled Beam (ACB) System Design

Figure 1: Schematic diagram of energy conserving Active Chilled Beam system

The ACB system consists of:

Air Handling Unit: Supply fan, air mixer, and cooling coils
Water Chilling Unit: Compressor, water pump, condenser, evaporator, and expansion valve
Control System:
- Dynamic prediction models for energy and environmental parameters
- PMV-based occupant comfort calculation
- Optimization model using improved Firefly Algorithm
- Optimal actuator control for ACB components

Energy consumption is modeled as: E = EC[(Ta, Hr), OF]

Where:

Ta = Air Temperature
Hr = Relative Humidity
OF = Operating Frequency (Fan, Pump, Compressor)

🤖 Machine Learning Model Development & Deployment

Dynamic Inference Model Architecture

Figure 2: Training and offloading of dynamic inference model to provide edge intelligence

🔧 Installation

Prerequisites

Raspberry Pi 4 Model B (or equivalent)
Ubuntu Server 20.04 LTS (preferred over Raspbian for better IoTivity support)
Python 3.8 or higher
Java Development Kit (JDK) 8 or higher
Git

📝 Setup Guide

Step 1: Raspberry Pi Configuration

1.1 Install Ubuntu OS on Raspberry Pi 4

# Download Ubuntu Server 20.04 LTS for Raspberry Pi
# Use Raspberry Pi Imager or balenaEtcher to flash the SD card

# Insert the SD card into Raspberry Pi and boot

Why Ubuntu instead of Raspbian?

Better IoTivity 2.2.2 support and compatibility
More comprehensive package repositories
Better remote development tools support
Enhanced security features

1.2 Initial Raspberry Pi Setup

# Update system
sudo apt-get update
sudo apt-get upgrade -y

# Install essential tools
sudo apt-get install -y git build-essential python3-pip
sudo apt-get install -y openssh-server

# Enable SSH (if not already enabled)
sudo systemctl enable ssh
sudo systemctl start ssh

# Note your Raspberry Pi's IP address
hostname -I

Step 2: Install Required Software on IoT Platform

2.1 Install PuTTY (For Remote Access)

Purpose: PuTTY initiates SSH commands allowing the IoT platform to send instructions remotely to the Raspberry Pi.

Installation (Windows):

# Download from: https://www.putty.org/
# Install PuTTY executable

# Alternative: Use Windows Terminal
# Windows Terminal also supports SSH natively
ssh ubuntu@<raspberry-pi-ip>

Connection Steps:

Open PuTTY
Enter Raspberry Pi IP address in "Host Name"
Port: 22
Connection Type: SSH
Click "Open"
Login with username and password

Other SSH Alternatives:

Windows Terminal (built-in SSH)
MobaXterm
Git Bash
PowerShell SSH client

2.2 Install WinSCP (For File Transfer)

Purpose: WinSCP is an FTP-based system used to transfer files from the platform to the device, including IoTivity 2.2.2 and Python scripts.

Installation:

# Download from: https://winscp.net/
# Install WinSCP

Configuration Steps:

Open WinSCP
File Protocol: SFTP
Host name: <Raspberry Pi IP address>
Port: 22
Username: ubuntu (or your username)
Password: <your password>
Click "Login"

Files to Transfer:

IoTivity source code
Python scripts
TFLite models
Configuration files

Step 3: Build and Install IoTivity on Device

3.1 Connect via PuTTY/SSH

# From your local machine
ssh ubuntu@<raspberry-pi-ip>

# Or use PuTTY with the IP address

3.2 Clone IoTivity Repository

# Clone IoTivity version 2.2.2 from GitHub
git clone --recursive --depth 1 --single-branch --branch 2.2.2 \
https://github.com/iotivity/iotivity-lite.git iotivity-222

cd iotivity-222

3.3 Build IoTivity

# Install dependencies
sudo apt-get install -y \
    build-essential \
    git \
    scons \
    libboost-dev \
    libboost-program-options-dev \
    libboost-thread-dev \
    uuid-dev \
    libssl-dev \
    libtool \
    libglib2.0-dev

# Build IoTivity server
./build-server-lite.sh

3.4 Run Sample Server

# Run the IoTivity server
./run-server-lite.sh

Expected Output:

Starting IoTivity Server...
OCF Server Started Successfully
Listening on port 5683
Waiting for client connections...

Step 4: IoT Platform Setup (Eclipse IDE)

4.1 Install Eclipse IDE

# Download Eclipse IDE for Java Developers
# https://www.eclipse.org/downloads/

# Install Eclipse

4.2 Configure Remote Systems in Eclipse

Purpose: Remote systems allow compiling and executing Java code remotely on the IoT device using remote compiler.

Configuration Steps:

Install Remote System Explorer
- Open Eclipse
- Go to Help → Install New Software
- Search for "Remote System Explorer"
- Install RSE (Remote System Explorer)
- Restart Eclipse
Create SSH Connection
- Open Remote Systems perspective
- Right-click in Remote Systems view → New Connection
- Select Linux as operating system
- Click Next
Enter Connection Details
- Host name: <Raspberry Pi IP address>
- Connection name: RPi-IoT-Device
- Description: Raspberry Pi IoT Device
- Click Next
Configure Services
- Select ssh.files (for file transfer)
- Select ssh.shells (for remote terminal)
- Select ssh.terminals (for remote execution)
- Click Next
Set Authentication
- User ID: ubuntu
- Password: <your password>
- Check "Save password"
- Click Finish
Test Connection
- Right-click on the connection → Connect
- Expand Sftp Files → Root to browse files
- Right-click → Launch Terminal to open remote shell

4.3 Configure Java Build Path

# In Eclipse, configure the IoTivity libraries
# Project → Properties → Java Build Path → Libraries
# Add External JARs → Browse to IoTivity Java bindings

4.4 Open SSH Connection and Run IoTivity

Steps:

In Eclipse, open Remote Systems view
Expand your Raspberry Pi connection
Right-click → Launch Terminal
Navigate to IoTivity directory:

   cd ~/iotivity-222

Start IoTivity server:

   ./run-server-lite.sh

In Eclipse, run the OCF Client Java code
Observe client-server communication in the terminal

Step 5: Install Python Dependencies

# On Raspberry Pi (via SSH)
pip3 install --upgrade pip
pip3 install tensorflow-lite
pip3 install flask
pip3 install numpy
pip3 install pandas
pip3 install mysql-connector-python

# Install additional requirements
pip3 install -r requirements.txt

Step 6: Transfer Files Using WinSCP

Files to Transfer:

Python Scripts:
- ocf_server.py
- flask_server.py
- sensor_drivers.py
TFLite Models:
- energy_model.tflite
- environment_model.tflite
- pmv_model.tflite
Configuration Files:
- config.yaml
- iotivity_config.json

Transfer Process:

Open WinSCP and connect to Raspberry Pi
Navigate to target directory on RPi (e.g., /home/ubuntu/iot-system/)
Drag and drop files from local system to RPi
Verify file permissions:

   chmod +x *.py
   chmod +x *.sh

Step 7: Configure MySQL Database

# Install MySQL on IoT Platform
sudo apt-get install mysql-server

# Secure installation
sudo mysql_secure_installation

# Create database
mysql -u root -p

CREATE DATABASE smart_home_db;
USE smart_home_db;
SOURCE schema.sql;

# Create user for application
CREATE USER 'iot_user'@'%' IDENTIFIED BY 'your_password';
GRANT ALL PRIVILEGES ON smart_home_db.* TO 'iot_user'@'%';
FLUSH PRIVILEGES;

🚀 Usage

Starting the Complete System

On Raspberry Pi (IoT Device):

# Terminal 1: Start OCF Server
cd ~/iot-system
python3 ocf_server.py

# Terminal 2: Start Flask Server
python3 flask_server.py

On IoT Platform:

# Terminal 1: Start OCF Client (Eclipse or command line)
cd aiot-platform
java -jar ocf-client.jar

# Terminal 2: Start Optimization Engine
python optimization_engine.py

Final Output of Client-Server Communication

Expected Console Output:

┌─────────────────────────────────────────────┐
│         OCF CLIENT-SERVER OUTPUT            │
├─────────────────────────────────────────────┤
│ [CLIENT] Discovering OCF devices...         │
│ [CLIENT] Device found: RPi-IoT-Device       │
│ [CLIENT] Connecting to: coap://192.168.x.x  │
│                                             │
│ [SERVER] Client connected from 192.168.x.x  │
│ [SERVER] Sending resource information...    │
│                                             │
│ [CLIENT] Received environmental data:       │
│   ├─ Temperature: 18.5°C                    │
│   ├─ Humidity: 51.96%                       │
│   ├─ CO2: 445.66 ppm                        │
│   └─ Energy: 2.52 kWh                       │
│                                             │
│ [CLIENT] Requesting predictions...          │
│ [SERVER] Running TFLite models...           │
│ [SERVER] Predictions complete               │
│                                             │
│ [CLIENT] Received predictions:              │
│   ├─ Predicted Temp: 19.2°C                 │
│   ├─ Predicted Humidity: 53.4%              │
│   ├─ Predicted Energy: 2.8 kWh              │
│   └─ PMV: 0.1 (Comfort)                     │
│                                             │
│ [OPTIMIZATION] Running Firefly Algorithm... │
│ [OPTIMIZATION] Optimal control found:       │
│   ├─ Fan Frequency: 37 Hz                   │
│   ├─ Pump Frequency: 40 Hz                  │
│   └─ Compressor Frequency: 43 Hz            │
│                                             │
│ [CLIENT] Sending control commands...        │
│ [SERVER] Actuators updated successfully     │
│                                             │
│ ✓ Energy Saving: 28.5%                      │
│ ✓ Comfort Index: 0 (Neutral/Comfortable)    │
│ ✓ Response Time: 4.8s                       │
│ ✓ Round Trip Time: 3.2ms                    │
└─────────────────────────────────────────────┘

📱 Mobile Application Interface

Android Client for User Interaction

Figure 3: Android client for getting occupant preference


## 📊 Results

**Key Findings:**
- Maximum energy savings: **36.82%** (at α = 1.0)
- Average improvement over baseline FA: **7%**
- Best balanced performance: **30.3%** savings at α = 0.5

### OCF Server Endpoints

#### 1. Get Environmental Data
```http
GET coap://[device-ip]:5683/oic/rd?rt=oic.wk.rdpub

Response:

{
  "temperature": 18.5,
  "humidity": 51.96,
  "CO2": 445.66,
  "energy_consumption": 2.52,
  "timestamp": "2023-10-15T14:30:00Z"
}

2. Get Predictions

GET coap://[device-ip]:5683/predict

Response:

{
  "temperature_prediction": 19.2,
  "humidity_prediction": 53.4,
  "energy_prediction": 2.8,
  "pmv": 0.1,
  "comfort_index": "neutral"
}

3. Update Actuator Control

POST coap://[device-ip]:5683/actuator/control

Request Body:

{
  "fan_frequency": 37,
  "pump_frequency": 40,
  "compressor_frequency": 43
}

Response:

{
  "status": "success",
  "message": "Actuators updated successfully",
  "updated_at": "2023-10-15T14:31:00Z"
}

Flask REST API

1. Get Current Status

GET http://[device-ip]:5000/status

Response:

{
  "device_status": "online",
  "ocf_server": "running",
  "models_loaded": true,
  "last_update": "2023-10-15T14:30:00Z"
}

2. Get Optimization Results

POST http://[server-ip]:8080/optimize

Request Body:

{
  "current_temperature": 18.5,
  "current_humidity": 51.96,
  "user_preference": 0.5,
  "constraints": {
    "temp_min": 20,
    "temp_max": 27,
    "humidity_min": 45,
    "humidity_max": 65
  }
}

Response:

{
  "optimal_control": {
    "fan_frequency": 37,
    "pump_frequency": 40,
    "compressor_frequency": 43
  },
  "predictions": {
    "energy_saving": 28.5,
    "comfort_index": 0,
    "predicted_temperature": 19.2,
    "predicted_humidity": 53.4
  },
  "algorithm": "improved_firefly",
  "iterations": 245,
  "convergence_time": 1.2
}

📚 Publication

Paper Details

Title: An OCF-IoTivity enabled smart-home optimal indoor environment control system for energy and comfort optimization

Authors: Anam Nawaz Khan, Atif Rizwan, Rashid Ahmad, Do Hyeun Kim

Journal: Internet of Things

Volume: 22

Article Number: 100712

Year: 2023

Publisher: Elsevier

DOI: 10.1016/j.iot.2023.100712

License: CC BY-NC-ND 4.0

📖 Citation

APA Citation

Khan, A. N., Rizwan, A., Ahmad, R., & Kim, D. H. (2023). An OCF-IoTivity 
enabled smart-home optimal indoor environment control system for energy 
and comfort optimization. Internet of Things, 22, 100712. 
https://doi.org/10.1016/j.iot.2023.100712

BibTeX Citation

@article{khan2023ocf,
  title={An OCF-IoTivity enabled smart-home optimal indoor environment control system for energy and comfort optimization},
  author={Khan, Anam Nawaz and Rizwan, Atif and Ahmad, Rashid and Kim, Do Hyeun},
  journal={Internet of Things},
  volume={22},
  pages={100712},
  year={2023},
  publisher={Elsevier},
  doi={10.1016/j.iot.2023.100712},
  issn={2542-6605},
  url={https://www.sciencedirect.com/science/article/pii/S2542660523000355}
}

🔗 Links

Publisher: Elsevier - Internet of Things
DOI: https://doi.org/10.1016/j.iot.2023.100712
PDF: Download from Publisher

🌐 Languages

Java: 51.8% (OCF Client, Eclipse IDE integration)
Python: 48.2% (Optimization, ML models, Flask server)

🤝 Contributing

We welcome contributions! Please see our Contributing Guidelines.

Development Setup

Fork the repository
Clone your fork
Create a feature branch
Make your changes
Test thoroughly
Submit a pull request

🐛 Troubleshooting

Common Issues

1. IoTivity Build Fails

# Ensure all dependencies are installed
sudo apt-get install -y build-essential git scons \
    libboost-dev libboost-program-options-dev \
    libboost-thread-dev uuid-dev libssl-dev

# Clean and rebuild
cd iotivity-222
./clean.sh
./build-server-lite.sh

2. PuTTY Connection Refused

# On Raspberry Pi, ensure SSH is running
sudo systemctl status ssh
sudo systemctl start ssh

# Check firewall
sudo ufw allow 22

3. WinSCP Transfer Failed

Verify network connectivity
Check file permissions on Raspberry Pi
Ensure sufficient storage space
Try SFTP protocol instead of SCP

4. Eclipse Remote Connection Issues

Verify SSH credentials
Check network firewall settings
Ensure RSE plugins are properly installed
Try reconnecting: Right-click → Disconnect → Connect

🙏 Acknowledgments

This work was supported by:

Institute for Information & communications Technology Promotion (IITP) Republic of Korea
- Grant NO. 2022-0-00980: Cooperative Information Intelligence Framework
- Grant NO. 2021-0-00188: Open source development for AI-enabled IoT platforms
Jeju National University - Mobile Computing Lab

📜 License

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Made with ❤️ for sustainable smart homes

Name		Name	Last commit message	Last commit date
Latest commit History 15 Commits
Docs/figures		Docs/figures
Energy_Optimization_SmartHome		Energy_Optimization_SmartHome
IoTivity_SmartHome_CoAP Connections		IoTivity_SmartHome_CoAP Connections
README.md		README.md

Ana119/Smart-Home-Optimal-Indoor-Environment-Control-System

Folders and files

Latest commit

History

Repository files navigation

OCF-IoTivity Smart Home Energy & Comfort Optimization System

🌟 Key Achievements

🎯 Overview

✨ Features

Core Capabilities

🏗️ System Architecture

Active Chilled Beam (ACB) System Design

🤖 Machine Learning Model Development & Deployment

Dynamic Inference Model Architecture

🔧 Installation

Prerequisites

📝 Setup Guide

Step 1: Raspberry Pi Configuration

1.1 Install Ubuntu OS on Raspberry Pi 4

1.2 Initial Raspberry Pi Setup

Step 2: Install Required Software on IoT Platform

2.1 Install PuTTY (For Remote Access)

2.2 Install WinSCP (For File Transfer)

Step 3: Build and Install IoTivity on Device

3.1 Connect via PuTTY/SSH

3.2 Clone IoTivity Repository

3.3 Build IoTivity

3.4 Run Sample Server

Step 4: IoT Platform Setup (Eclipse IDE)

4.1 Install Eclipse IDE

4.2 Configure Remote Systems in Eclipse

4.3 Configure Java Build Path

4.4 Open SSH Connection and Run IoTivity

Step 5: Install Python Dependencies

Step 6: Transfer Files Using WinSCP

Step 7: Configure MySQL Database

🚀 Usage

Starting the Complete System

On Raspberry Pi (IoT Device):

On IoT Platform:

Final Output of Client-Server Communication

📱 Mobile Application Interface

Android Client for User Interaction

2. Get Predictions

3. Update Actuator Control

Flask REST API

1. Get Current Status

2. Get Optimization Results

📚 Publication

Paper Details

📖 Citation

APA Citation

BibTeX Citation

🔗 Links

🌐 Languages

🤝 Contributing

Development Setup

🐛 Troubleshooting

Common Issues

1. IoTivity Build Fails

2. PuTTY Connection Refused

3. WinSCP Transfer Failed

4. Eclipse Remote Connection Issues

🙏 Acknowledgments

📜 License

About

Resources

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Languages

Packages