Activity-based Profiling for Energy Harvesting Estimation (2305.08025v1)

Abstract: We propose a novel activity-based profiling framework to estimate IoT users' harvested energy based on their daily activities. Energy is harvested from natural sources such as the kinetic movement of IoT users. The profiling framework captures the users' physical activity data to define activity-based profiles. These profiles are utilized to estimate the harvested energy by IoT users. We train and evaluate our framework based on a real Fitbit dataset.