Abstract: With the rapid development of mobile crowdsourcing,crowdsourcing programs in the market have sprung up.They distribute tasks and use the power of the crowd to perform the tasks for collecting data and an effective incentive mechanism in mobile crowdsourcing becomes very important.However,the existing incentive mechanisms nowadays partially consider the reputation value,location and execution time of workers,which makes it difficult for crowdsourcing platform to select high-quality workers and assign multiple tasks on limited budgets or other constraints.To solve the above problems,this paper proposes an incentive mechanism on the basis of the multi-constrained worker selection (MSIM),which relies on two related algorithms.One is the algorithm of worker selection based on improved reverse auction model,which comprehensively considers many important limitations to select great workers to perform the tasks,such as worker reputation,geographical location,task completion degree and result quality.The other is the algorithm of reward and punishment by evaluation,which contains the evaluation of task-perceiving results and workers' reputation.The experimental results showed that not only can MSIM select excellent workers,but also it improved the credibility of the task results and the reputation of workers.It is proved within this paper that the MSIM is an effective incentive mechanism.

Key words: Mobile crowdsourcing, Worker selection, Multiple constraints, Result evaluation, Incentive mechanism