Energy-Aware Service Composition of Configurable IoT Smart Things

This paper presents a three-tier framework to facilitate the composition of Internet-of-Things (IoT) services, where these IoT services represent functionalities provided by heterogenous smart things. Various IoT services are categorized into service classes through the categorization of their functionalities. A service network is constructed by considering the invocation relationship between service classes, and service class chains are generated using traditional Web service composition techniques to satisfy the requirement from the functional perspective only. Considering the factors, including spatial and temporal constraints, energy efficiency, and the functional configurability, IoT service composition can be reduced to a multi-objective optimization problem. Heuristic algorithms, such as genetic algorithm (GA), ant colony optimization (ACO), and particle swarm optimization (PSO), are adopted to search for optimal IoT service compositions. Experimental results show that PSO performs better than GA and ACO in searching for approximately optimal IoT service compositions and reducing the energy consumption of smart things in the network.