This paper presents the design, development, and evaluation of an IoT-enabled wearable sensor system for real-time health and posture monitoring, aimed at enhancing user safety and performance in both athletic and smart-city mobility contexts. The system integrates two prototypes: a posture-correction vest employing dual MPU6050 inertial measurement units with an on-body buzzer for immediate feedback, and an activity-monitoring glove equipped with an MPU6050 for step detection, a pulse-rate sensor for heart-rate monitoring, and Wi-Fi connectivity for cloud-based data logging via ThingSpeak. Data are processed locally on an Arduino platform and analyzed using a MATLAB-based artificial neural network employing the Levenberg–Marquardt algorithm to generate personalized warm-up or rest recommendations based on heart-rate thresholds. Experimental evaluation demonstrated that the posture-correction vest achieved a sensitivity of 85.5% in detecting poor posture, while the glove reached accuracies of 96.52% and 94.74% for heart-rate monitoring and step counting, respectively. The proposed system offers a low-cost, scalable, and extensible platform that combines real-time feedback with remote analytics, making it suitable for deployment in sports training, occupational health, and smart-city mobility applications.
Mohammed et al. (Thu,) studied this question.