Energy consumption optimization based on Endirichlet distribution in Mobile crowd sensing

In this paper, we consider a strategy for classifying tasks based on perceptual attributes and selecting participants based on perceptual attributes. In the covert of the classification task using the Latent Dirichlet Allocation (LDA) distribution, defined tasks according to the function of using sensors required for all kinds of attributes, the MCS system by sensing properties and energy consumption for the choice of participants, the participants set can meet the task of all sensory attributes, Our goal is to select participants to improve task completion while reducing overall system energy consumption. According to the perceptual attributes of tasks and the relationship between participants' energy consumption and task completion rate, a heuristic algorithm is proposed to solve these problems. Simulation results show that this algorithm is superior to other algorithms.