Edge computing task unloading dynamic pricing method based on context dobby

The invention relates to an edge computing task unloading dynamic pricing method based on a context dobby, and belongs to the technical field of edge computing. According to the method, firstly, online unloading information of users and online income information of service providers are collected according to a multi-user multi-task edge computing dynamic environment, a pricing problem is modeled as a context dobby model, and multivariate Gaussian distribution is constructed according to the collected context user unloading information; an online edge service price is selected by adopting a Thompson sampling method, namely, a service provider is induced to perform price selection by applying Bayesian posteriori, and corresponding parameters are updated through each round of online reward income. According to the method, modeling is carried out on the edge calculation dynamic pricing problem based on the dobby model, the purpose of maximizing the income of a service provider is achieved through online pricing