A Survey of Decision-Theoretic Models for Cognitive Internet of Things (CIoT)

Abstract

Communication technology and wide spread usage of Internet of Things (IoT) are rapidly becoming the core enabler for consumers to use their smart devices in their daily routine. These smart devices are gradually transforming the IoT scenario into a new paradigm called cognitive Internet of Things (CIoT). We believe that CIoT will revolutionize many service sectors including smart cities, transportation, healthcare, and environmental monitoring. The current development in CIoT is still elusive as it has to achieve effective interoperability and autonomous decision making. Most importantly, the decision theoretic models are seen as an enabler to achieve effective interoperability among heterogeneous CIoT objects. In this paper, we provide a survey of the existing decision theoretic models and their usage for CIoT, an architectural CIoT framework is also proposed to discuss the open issues and solution for potential challenges emerging in the area of CIoT research.

Keywords

• Cognitive Internet of Things (CIoT)
• decision theoretic models
• game theory
• multi-agent learning
• Markov decision process
• multi-arm bandit problem