An adaptive immune algorithm for service-oriented agricultural Internet of Things

Abstract Internet of Things (IoT) has the characteristics of multi-source, multi-type and unequal amount of task. Considering multiple requests at stochastic moment in the agricultural IoT service, an optimization model of minimum service cost is built in this paper, and an adaptive immune algorithm with endocrine regulation (AIE) is proposed to solve the problem of dynamic service for multiple requests. Based on this model, two service providing strategies are put forward, i.e. single service and collaborative service. The former means that one request task is completed by one service provider, while the latter means that one request task is collaboratively completed by multiple service providers. The service cost and service time are respectively evaluated under two service providing strategies. The simulation results verify the feasibility and effectiveness of the proposed AIE through comparison to the immune algorithm without endocrine regulation. Meanwhile, it is demonstrated that the proposed AIE can obtain more excellent performance than genetic algorithm and particle swarm optimization.