Design and Implementation of Real-time Embedded Data Acquisition and Classification with Web-based Configuration and Visualization

This research proposes some useful and practical techniques to design and implement the real-time embedded data acquisition and classification. A web-based used as graphical user interface is also provided to visualize data, parameters and behaviors of the proposed system. A low-cost microcontroller is chosen to run special bootloader and application program. The application program is designed and implemented using non-blocking event-driven real-time operating system. It reads data from sensors, performs K-means classification algorithm, and sends data and system states to the web-based application running on a web browser. In addition, the microcontroller system is designed to support many IoT applications. It is composed of all basic inputs and outputs, e.g., ADCs, PWMs, Wi-Fi, USB-to-UART and data storage. The users can visualize the measured data and examine behaviors of the K-means algorithm on the web browser in real-time. The web-based application is designed and implanted using new technology of web-based application development techniques. It can be run on any device that has web browser. The microcontroller and web-based application can exchange their data over internet network using HTTP, MQTT or WebSockets protocol. It supports serial port communication as well. The experimental results show that the proposed system and algorithms running on the microcontroller and web browser can perform data acquisition and classification in real-time manner and it can be used in real-world applications as expected.