Mobile Healthcare Systems with Multi-cloud Offloading

The fast growth of cloud computing has attracted more companies to migrate their in-house IT applications into cloud and it also occurs in the medical field. A mobile healthcare system with cloud offloading is considered in this paper and it can be divided into two stages: sensor network and cloud offloading. In the first stage, information collected by body sensors should be transmitted to a remote mobile device. In order to save energy, an energy-efficient transmission scheme called cooperative multi-input multi-output (MIMO) is constructed for the data transfer when allowing individual sensor nodes to cooperate with each other. In the second stage, two offloading schemes called self-reliant multi-cloud offloading system and multi-cloud offloading system are proposed and further analyzed based on serve topology and optimal graph partition. The former provides stability but with high communication cost, while the latter reduces communication cost but is less stable. Both schemes can be applied to other scenarios in which we would like to perform offloading on multiple servers.