A fog-based ubiquitous exercise healthcare monitoring framework for smart cities

Fog-based exercise healthcare monitoring (EHM) has attracted many interests in the field of smart cities (SCs) as an effective supplement to the cloud-based applications to reduce the transmission delay and data quantity swarming from smart health devices toward the Internet. However, due to the amount of heterogeneous data from a variety of Internet of Things (IoT) devices, providing fast-responsive or low-latency service has always been a great challenge in EHM. Hence, to address these challenges, we propose a fog-based framework for ubiquitous exercise monitoring capable of efficiently transmitting and processing data at the edge of the network. For this framework, a deep learning-based traffic prediction method is proposed to ensure optimal network performance. The simulation results demonstrate that our proposed framework has better network performance, in terms of real-time, average completion time and lost rate of packet queue, compared with several state-of-the-art frameworks.