Research On User Incentive And Optimal Selection Method In Mobile Crowd Sensing

In recent years,the popularity of mobile devices and the growing demand for smart perception in cities have led to an alternative or supplementary urban perception method,called mobile crowd sensing.In terms of data collection,any individual participant cannot provide enough sensing samples.The mobile crowd sensing system can obtain high-quality,high-coverage sensing results by gathering the sensing data of a large number of participants.Therefore,motivating a large number of sensing users to join the sensing system is a prerequisite for good results.However,most researches related to mobile crowd sensing assume that a large number of sensing users exist in the sensing system,and it is particularly important to analyze user recruitment and task diffusion.When sensing users participate in sensing tasks,the submitted sensing data varies from good to bad.In order to ensure the common profits of sensing users and the system platform,it is more important to encourage users to provide high cost-effective sensing quality.When the system allocates tasks,due to the self-interest of mobile users,there may be a surplus of mobile users in some task areas,and mobile users in some areas are too sparse,resulting in the completion of the task cannot be guaranteed when the number of mobile users is sufficient.Therefore,this paper conducts research from the aspects of user participation,system social welfare and user optimal selection.The main work of this paper is as follows:1.Aiming at the problem that the mobile crowd sensing system relies on a specific platform with a large user group presupposed,a method of perceiving user diffusion analysis based on knowledge graph reasoning is proposed.Using social networks as a recruitment platform,combined with the knowledge graph to establish a knowledge graph for the mobile crowd sensing system,filter and filter to obtain effective social user attributes,calculate the matching degree between users and tasks,and deduce user influence by rules to get the user's society benefit.Select social users who have sufficient contribution value to the platform and meet the threshold of willingness to participate as candidate users of the mobile crowd sensing system,thereby increasing the system user group.2.Aiming at the problem of large differences in the quality of user submissions of the mobile crowd sensing system,a social welfare-oriented MCS social user incentive mechanism is designed.Consider motivating users to participate under the constraints of a limited budget.Focusing on socially aware users represented by self-organizing social networks,with the goal of maximizing social welfare,a reverse auction mechanism is proposed.The winning bidder is selected according to the net contribution margin,reducing the sensing data redundancy,and the winning bidder is paid based on the critical price.Through the implementation of the incentive mechanism,the system platform and users can obtain the maximum benefits.3.For the optimal selection strategy of mobile users in the mobile crowd sensing system,a quantum-inspired firefly algorithm and particle swarm optimization method are proposed to optimize resource allocation for mobile users from the aspect of task coverage.First,use quantum computing to initialize the mobile user population;secondly,calculate the fitness function designed according to the objective function,and compare and sort them;then perform the corresponding quantum motion according to the comparison result of the fitness value;finally update the population after the quantum motion,and judge whether the termination condition is reached.In this process,the optimal mobile user group is solved to ensure the coverage and completion rate of sensing tasks,thereby improving the aggregation results of sensing data.