SAMuS: service-oriented architecture for multisensor surveillance in smart homes

Area of Science:

• Computer Science
• Artificial Intelligence
• Internet of Things

Background:

• Smart home surveillance systems require integrated solutions for effective multisensor data utilization.
• Existing architectures often lack dynamic sensor composition and automatic deployment capabilities.

Purpose of the Study:

• To design a service-oriented architecture for multisensor surveillance in smart homes.
• To enable automatic deployment, dynamic selection, and composition of sensors.
• To enhance complex event detection through sensor collaboration.

Main Methods:

• Developed a service-oriented architecture with Web-connected sensors and a uniform Web API.
• Utilized RESTdesc for sensor description and a novel solution for automatic Web API composition compatible with Semantic Web reasoners.
• Implemented a smart Kinect sensor with dynamic IR/RGB switching and integrated pressure sensor feedback.

Main Results:

• Demonstrated successful automatic composition of Web APIs for sensor integration.
• Showcased enhanced person detection through collaboration between Kinect and pressure sensors.
• Validated platform scalability with composition times under a few hundred milliseconds for numerous Web APIs.

Conclusions:

• The proposed service-oriented architecture effectively integrates multisensor data for smart home surveillance.
• Automatic sensor composition and collaboration significantly improve complex event detection.
• The platform demonstrates robust scalability for real-world smart home applications.