Availability Modeling and Performance Improving of a Healthcare Internet of Things (IoT) System

• Published in: IoT 2021-06, Vol.2 (2), p.310-325
• Main Authors: Tang, Shensheng, Xie, Yi
• Format: Article
• Language: English
• Subjects: Access control; availability modeling; Biosensors; Body temperature; Business metrics; Cloud computing; Communication; degraded service; Failure; full service; Global positioning systems; GPS; Heart rate; Human mechanics; Infrastructure; Internet of Things; IoT; Markov process; Network topologies; Neuromuscular diseases; Performance evaluation; Physiology; Sensors; Simulation; Software; stochastic modeling; Technology; Wireless networks
• ISSN: 2624-831X; 2624-831X
• DOI: 10.3390/iot2020016

Internet of Things (IoT) is improving human life in a more convenient and simpler way. One of the most promising IoT applications is healthcare. In this paper, an availability model of a healthcare IoT system is proposed which is composed of two groups of structures described by separate Markov state-space models. The two separate models are analyzed and combined to implement the whole IoT system modeling. The system balance equations are solved under a given scenario and some performance metrics of interest, such as probabilities of full service, degraded service, and the system unavailability, are derived. Detailed numerical evaluation of selected metrics is provided for further understanding and verification of the analytic results. An availability performance improving (API) method is also proposed for increasing the probability of system full service and decreasing the system unavailability. The proposed system modeling and performance improving method can serve as a useful reference for general IoT system design and evaluation.