Research On Participant Selection Strategy In Group Intelligence Perception Network Based On Time Window

The rapid development of wireless networks and intelligent devices has given birth to a new type of perception network,namely mobile swarm intelligence perception network.Mobile group-intelligence perception network takes a large number of participants carrying smart devices as perception nodes,but these participants usually do not participate in perception tasks for free.Therefore,how to select appropriate participants has always been the key field of mobile group-intelligence perception network research.Some perceptual tasks in mobile group-intelligence perception network require participants to collect continuous perceptual data over a period of time.This paper studies the selection strategies of mobile group-intelligence perception network participants of this task type,and the main contents are as follows:This paper proposes a time window incentive mechanism for participants.The tasks released by mobile group-intelligence perception platform have data integrity requirements,and the tasks released have a certain period of time.Based on specific perception of the time,put forward the prize pool mechanism,for participants to perceive time less high rewards to part to attract participants,in task optimization algorithm based on dynamic programming is the choice of participants,and then puts forward an additional incentive mechanism^,according to each of the participants involved in time for different incentive strategy.Through experimental verification,the incentive mechanism proposed in this paper has a significant effect on increasing the number of participants,increasing the amount of prize pool and the remuneration of participants.A time window-dependent participant selection strategy is proposed.This mechanism mainly includes:design a participant selection method based on dynamic programming algorithm,the target for the selected participants perceive time cover task period benefit maximize data at the same time,it also joined the participants XinYuZhi update strategy,depending on the degree of willingness to participants involved in the task and update participants XinYuZhi data quality.Through experimental verification,the participant selection strategy proposed in this paper,compared with MST and Random,has a better effect on data reliability,data efficiency,perceived cost and other aspects.The proposed incentive mechanism ensures that there will be enough participants to participate in the perceptual task,and the participant selection strategy is to select appropriate participants from enough participants to meet the task requirements.These two parts of work provide a complete work mode for the task platform.