Research On The Preferred Method Of Mobile Crowd-sensing Multitasking Participants

With the advent of the Internet of Things era,various portable devices are used more and more widely in the crowd,and the demand for intelligent terminals is also growing,which makes mobile terminal devices have more and more powerful functions of perception,calculation and storage.Crowd sensing comes into being.Its characteristic is not only to use traditional sensor networks,but also to use mobile terminal devices to collect data.As a new service mode to perceive the environment,collect data and provide information for payment,it has gradually become one of the research hotspots.The incentive mechanism is one of the main methods to realize the model.Crowd sensing needs incentive mechanism to encourage participants to actively join the crowd sensing system.Moreover,while collecting enough high-quality data and information,the expenditure cannot exceed the platform budget.In this paper,we will study the optimization methods of multi-task participants in mobile crowd sensing.The specific research work of this paper includes:(1)Under the condition of spatial coverage in a certain perception area,based on the task-centered VT-MOST algorithm,the past speed of participants is used to estimate the value of participants when performing tasks,and the incentive payment is minimized by minimizing the moving distance of participants.Under the condition that the cost satisfies the constraint condition,the candidate with greater utility is selected as the participant to perform the task set,and the estimation of the participant set is minimized.(2)The whole perception area is divided into a group of grids by using clustering idea,and the candidate participants and tasks in each grid are classified and matched based on user-centered PT-MOST algorithm.According to the mobile mode,the participants with higher utility are selected to reduce the computational complexity of participants' task allocation.It effectively enhances the minimization of moving distance in VPT-MOST algorithm.(3)The credibility value of participants is used as an evaluation index to measure the reliability of information collected by participants.By dividing regions,the subregion with the largest number of tasks in each subregion is selected,and the candidate with the largest credibility value in its subregion is selected as the participant's V-MOST algorithm,and the user credibility update function is set.And effectively reduces the cost of the platform.Maximize the benefits of the platform.