Design and Implementation of SEMAR IoT Server Platform with Applications.

Area of Science:

• Computer Science
• Environmental Science
• Engineering

Background:

• Internet of Things (IoT) technologies are rapidly advancing, enabling smart city development.
• Current IoT application systems often lack standardization, hindering integration and collaboration.
• Independent deployment of IoT systems leads to interoperability issues in smart city environments.

Purpose of the Study:

• To design and implement a standardized IoT server platform for integrating diverse IoT application systems.
• To develop a platform, SEMAR, that supports Big Data environments with machine learning capabilities.
• To enhance interoperability and flexibility in smart city IoT integrations.

Main Methods:

• Developed the Smart Environmental Monitoring and Analytical in Real-Time (SEMAR) IoT server platform.
• Integrated built-in functions for data aggregation, synchronization, classification, and machine learning.
• Implemented plug-in functions and Representational State Transfer Application Programming Interface (REST API) services for data access and external integration.
• Integrated five distinct IoT application systems for evaluation: air-conditioning, indoor localization, water quality, environment, and air quality monitoring.

Main Results:

• SEMAR successfully integrated five different IoT application systems.
• The platform demonstrated efficient data aggregation, synchronization, and classification using machine learning.
• SEMAR provided real-time data access through user interfaces, text files, and REST API services.
• Evaluated systems showed direct data acceptance from various sensors and network connections.

Conclusions:

• The proposed SEMAR IoT server platform offers superior flexibility and interoperability compared to existing solutions.
• SEMAR facilitates seamless integration of IoT device management, data communication, and decision-making functions.
• The platform's design allows easy integration with external programs and IoT applications without code modification, confirming its effectiveness and efficiency.