Multimodal Sensor Fusion-Based Lightweight Pedestrian Wearable System for Continuous Health Monitoring and Location Tracking

Real-time vital signs monitoring plays a crucial role in national security, health management, smart healthcare, and emergency response. However, current health monitoring systems generally lack indoor positioning capabilities and rely solely on GPS for outdoor navigation, potentially delaying the identification of the patient's precise location in an emergency and missing the optimal rescue window. To address this challenge, this paper develops a multimodal health monitoring and localization system that integrates a foot-mounted inertial sensor with a wrist-mounted health sensor. The proposed system leverages dead reckoning technology and a high-precision gait recognition algorithm based on the foot-mounted inertial sensor to achieve seamless and accurate three-dimensional pedestrian localization in both indoor and outdoor environments. Simultaneously, the wrist-worn device integrates electrocardiogram (ECG), photoplethysmography (PPG), and accelerometer sensors, enabling real-time monitoring of vital signs and fall detection. Experimental results demonstrate that the system effectively monitors physiological parameters such as blood pressure, blood oxygen saturation, heart rate, stress levels, and potential chronic diseases. The gait pattern recognition model based on error-correcting output codes (ECOC) achieves an accuracy of over 96%, precisely distinguishing various gait behaviors, including walking, running, stair ascent/descent, elevator usage, and falls. This wearable system is lightweight and easy to use, making it promising for applications in medical monitoring, location tracking, and emergency rescue, providing a more accurate and efficient solution for real-time health monitoring and emergency response.