IoT Architectures for Noninvasive Blood Glucose and Blood Pressure Monitoring

Continuous noninvasive blood glucose (sugar) and blood pressure (BP) monitoring using Photoplethysmogram (PPG) is an important research area. As the research in this area is in the beginning stages, there is a need to preserve the raw PPG signals for further analysis and improvement of the prediction models. The popular IoT data optimization techniques such as lossy data compression, extracting and transmitting the features and discarding the raw data are not applicable in this case. Transmitting and storing the massive raw PPG signals over the IoT infrastructure presents a major research challenge. Another research challenge is to ensure high availability of wearable health monitoring IoT edge device for user interaction irrespective of the network availability. In this paper we present an Adaptive IoT model which the existing cloud and fog computing architectures along with a new hybrid computing architecture to solve these research challenges. We also present our BP and sugar monitoring system which implements this Adaptive IoT model using the cloud, fog and hybrid computing architectures.