Research On Multi-Round Task Allocation Of Crowd Collaboration In Edge Cloud Environment

Mobile Crowd Sensing(MCS)is a new paradigm of data collection,analysis and sharing that uses a large number of crowd workers as the source of sensing and takes advantage of the wide distribution and data concentration of crowd workers to complete a series of sensing tasks.The MCS contains three main research areas: task allocation,quality control and privacy protection.Amongst other things,task allocation is a central issue in applied MCS research.Task allocation focuses on how to allocate budgets rationally to accomplish an effective match between crowd workers and tasks with limited budgets.Quality control is to ensure that the data uploaded by crowd workers is true and valid.Privacy protection is designed to protect crowd workers from sensitive information during the uploading of data.This paper addresses the issue of task allocation.At this stage of research on task allocation in MCS with limited budgets,most studies assume that crowd workers have the same level of completion quality for all tasks,ignoring the diversity of crowd workers’ abilities and failing to consider the true cost of crowd workers,resulting in poor overall quality of task completion and serious budget wastage.Most of the existing task allocation methods adopt one-time allocation,lacking the negotiation and communication process between crowd workers and task publishers on task requirements,resulting in overly rigid task requirements,which makes it difficult to maximize the overall social welfare of crowd workers and task publishers.To address the above problems this paper investigates a multi-round task allocation and incentive mechanism based on the edge cloud.The specific work is as follows:1.In order to solve the problem of single task allocation mechanism and improve task allocation efficiency.A framework of edge cloud-based MCS system is constructed,and a multi-round collaborative task allocation model(Mrc WS)under the edge cloud is proposed.The multi-round task allocation model divides the edge cloud region according to the density of crowd workers and recommends suitable tasks by predicting the trajectories of crowd workers.After a crowd worker accepts a task,the task is recommended twice by predicting the future motion trajectory,and the crowd worker is encouraged to complete the unassigned task by setting the appropriate task reward and reputation value.Finally,experimental validation using a real dataset,Foursquare,demonstrates that the proposed model is effective in improving task matching and data quality,and maintaining the longterm stability of the system.2.In order to solve the problem of overly rigid task demand due to one-time task assignment,and to maximize the revenue of MCS system.A multi-round negotiation task allocation and incentive mechanism based on social networks is proposed,which targets the multi-attribute negotiation of tasks.Firstly,a multi-round bargaining model between task publishers and crowd workers is developed to increase the interaction between task publishers and crowd workers.Secondly,the concept of information dissemination power is introduced to motivate crowd workers to disseminate information and improve task response rate.Finally,an experimental validation is conducted on the real dataset to demonstrate that the proposed model can effectively improve task matching and data quality,and maximise social welfare.