An Efficient Resource Allocation Approach Based on a Genetic Algorithm for Composite Services in IoT Environments

As various types of Internets of Things (IoT) are deployed in a wide range of areas, the need arises to utilize various IoT resources dynamically to accomplish user tasks. We call this environment an urban-scale IoT environment, where various IoT resources that are necessary to accomplish user tasks are directly connected to each other via users´ mobile devices, such as their smart phones. IoT resources are utilized as resources with which to run a composite service that supports user tasks. In this urban-scale IoT environment, it is essential to create efficient binding between a service and an IoT resource so as to execute a composite service for a task successfully. In this paper, we propose a service resource allocation approach which minimizes data transmissions between users´ mobile devices and which effectively deal with the constraints of these types of environments. We transformed the resource allocation problem into a variant of the degree-constrained minimum spanning tree problem and applied a genetic algorithm to reduce the time needed to produce a near-optimal solution. We also defined a fitness function and an encoding scheme to apply the genetic algorithm in an efficient manner. The proposed approach shows a 97% success rate on average when used to find near-optimal solutions. In addition, it takes significantly less time than the brute force approach.