Design and Implementation of SEMAR IoT Server Platform with Applications

Abstract

Nowadays, rapid developments of <i>Internet of Things (IoT)</i> technologies have increased possibilities of realizing <i>smart cities</i> where collaborations and integrations of various <i>IoT application systems</i> are essential. However, IoT application systems have often been designed and deployed independently without considering the standards of devices, logics, and data communications. In this paper, we present the design and implementation of the <i>IoT server platform</i> called <i>Smart Environmental Monitoring and Analytical in Real-Time (SEMAR)</i> for integrating IoT application systems using standards. <i>SEMAR</i> offers <i>Big Data</i> environments with <i>built-in</i> functions for data aggregations, synchronizations, and classifications with <i>machine learning</i>. Moreover, <i>plug-in</i> functions can be easily implemented. Data from devices for different sensors can be accepted directly and through network connections, which will be used in real-time for user interfaces, text files, and access to other systems through <i>Representational State Transfer Application Programming Interface (REST API)</i> services. For evaluations of <i>SEMAR</i>, we implemented the platform and integrated five IoT application systems, namely, the <i>air-conditioning guidance system</i>, the <i>fingerprint-based indoor localization system</i>, the <i>water quality monitoring system</i>, the <i>environment monitoring system</i>, and the <i>air quality monitoring system</i>. When compared with existing research on IoT platforms, the proposed <i>SEMAR</i> IoT application server platform offers higher flexibility and interoperability with the functions for IoT device managements, data communications, decision making, synchronizations, and filters that can be easily integrated with external programs or IoT applications without changing the codes. The results confirm the effectiveness and efficiency of the proposal.