A Wearable Internet of Things Based System with Edge Computing for Real-Time Human Activity Tracking

Wearable devices for human physical activity tracking are becoming a commonly used gadget in daily life. Commercial devices such as smartwatches and smart bands are currently available from many different vendors. Most of them are capable of providing health statistics through sensors embedded in them. However, they only provide the basic information in the form of a quantity, which does not have a direct clinical relevance. A major reason for this is the use of cloud computing by most of these devices. Cloud systems managed by commercial companies charge for their premium services, which allows users to get access to their data. Besides, healthcare regulatory bodies do not yet approve the analytics performed by cloud systems. This makes these systems unsuitable for home based physical therapy or rehabilitation monitoring. Leveraging on the Wearable Internet of Things (WIoT) and edge computing technologies, we developed a low cost, portable, and wearable device which can collect human activity data in real-time and transfer it to a local server for computing and analysis. We tested our device for two different activities and successfully transferred the data from tracking device to an edge computer at frequency of the order of 100 Hz.