Asynchronous Data Communication Architecture on Real-Time Face Recognition for Hybrid Laboratory System

The Internet of Things (IoT) and Artificial Intelligence (AI) have reshaped educational technology, increasing the chances of developing more effective and engaging learning environments. This study proposes an architecture for asynchronous data communication with real-time face recognition, a virtual and remote labs confluence. The architecture uses IoT hardware and software to comprise critical components such as IoT cameras, message brokers, FTP servers, and AI processing servers. The aim is to optimize resource usage and maximize response time efficiency. The architecture employs message queuing to manage high volumes of face image data, facilitating immediate verification and rapid access. Our findings demonstrate that the system can achieve face recognition with 91.7% accuracy and an average system communication latency of 2.631 seconds, effectively leveraging asynchronous communication to optimize resource usage and computational efficiency. The research comprehensively measures the data communication metrics and face recognition metrics. It also explains how asynchronous data communication can be strategically deployed in a hybrid lab environment to enhance educational and operational efficacy.