Research On Worker Recruitment Algorithm For Mobile Crowdsensing System

In recent years,with the development of mobile Internet and smart sensor technology,mobile crowd sensing computing paradigm has attracted wide attention in academia,industry,and business.Mobile crowd sensing system utilizes the sensing and computing capabilities of smart devices carried by workers to cooperate through mobile Internet to accomplish complex sensing tasks.In mobile crowd sensing system,worker recruitment is a core common research issue,and it is a combinatorial optimization problem that satisfies a variety of optimization objectives and constraints considering tasks,workers,and other factors.Research on worker recruitment problems in mobile crowd sensing faces the problem of cold start due to the insufficient number of workers participating in sensing tasks,the problem of balancing sensing quality and sensing cost due to single worker recruitment,the problem of not being adapt to real-time and large-scale mobile crowd sensing tasks and so on.In view of the above problems,this paper mainly studies the following three aspects:1)Worker recruitment on hybrid networks.This paper borrows the idea of influence propagation on social networks and proposes a worker recruitment algorithm based on hybrid networks mixing social networks and communication networks,which can solve the cold start problem to some extent.The core idea is that firstly select seed workers according to the recruitment probability by using communication networks,then initiate the task spread on social networks and communication networks at the same time in a greedy way.The goal of worker recruitment is to maximize task spatial coverage.When calculating the probability of recruitment,this paper considers various factors such as worker's ability,sojourn time,and worker's movement to improve the accuracy of recruitment probability.Experimental results show that compared with the existing algorithms,the algorithm proposed in this paper can guarantee the time constraint of the task and have better performance in terms of spatial coverage and running time.2)Worker recruitment with budget constraints.This paper proposes a two-phase worker recruitment framework,which recruits workers in two phases.It can embody a tradeoff between task quality and cost.First,in the offline phase,borrowing the idea of influence propagation on communication and social networks,this paper proposes offline opportunistic worker recruitment algorithm to recruit workers during their daily routines to complete tasks which can alleviate the cold start problem in traditional mobile crowd sensing system.Then,in the online phase,in order to reduce the computational complexity,this paper devises online participatory worker recruitment algorithm to incent workers to move to specific subareas obtained by subareas clustering to fulfill sensing tasks.In both phases,this paper considers guaranteeing the incentive cost and time constraint.Experimental results show that compared with other methods,the framework proposed in this paper has better performance in terms of spatial coverage and running time under budget constraints.3)Worker recruitment based on edge-cloud collaboration.This paper proposes a hierarchical worker recruitment framework based on edge-cloud collaboration,which can be applied to large-scale and real-time sensing tasks.The problem space is partitioned according to the sensing task scope in the cloud computing layer,while real-time data is processed and aggregated in the edge computing layer.Besides,the sensor configuration of sensing devices carried by workers,the quotation of sensing tasks from workers,the maximum number of tasks assigned to workers,worker's availability,and other factors are considered from the perspective of workers when recruiting workers.Then a mathematical model is constructed to judge whether workers can be recruited.Experimental results show that compared with existing methods,the proposed worker recruitment framework in this paper can achieve better performance in terms of spatial coverage and running time while ensuring the total cost and time constraints of tasks.