Incentive Mechanism For Mobile Crowd Sensing Based On Incomplete Information Game

Mobile crowd sensing(MCS)networks utilize the mobile devices(such as smartphones and wearable devices)as sensing nodes,and leverage them to collecting sensing data by collaborating with each other through the mobile Internet for completing large-scale and complex sensing tasks.Compared to traditional sensor networks,MCS networks have advantages of lower data collection costs,widespread of sensing range,easier system maintenance,better scalability and so on,which makes it a promising research filed and capable of many potential applications.MCS applications require a large group of mobile smartphone users' collective contribution in order to complete sensing tasks.However,while participating in a MCS task,smartphone users consume their own resources such as communication,battery and computing power.In addition,users also expose themselves to potential privacy threats by sharing their sensed data with location tags.Thus,smartphone users may not be willing to join a MCS system unless they receive enough compensation for their resource consumption.Therefore,a well-designed incentive mechanism to encourage adequate user participation is needed for MCS systems.The main work is as follows:1.To enhance the enthusiasm of user participation in MCS,modeling the MCS process and analyzing the participate user revenue function under incomplete information,and consider a more real MCS scenario where winners drop out of sensing tasks with random probability,a novel incentive mechanism for location-aware sensing that maximizes user's utility is proposed.The incentive mechanism contains two parts: winner selection and payment determination.In the first part,a task-centric winner selection algorithm is introduced and it can improve the sensing task coverage ratio.In this algorithm,for each sensing task,the user with small bids and large task values will be selected as the winner.The second part is a payment scheme to stimulate the continuous participation of users,which determines the payment to winners by a time proportional share rule to ensure the truthful of incentive mechanism,and the winners can obtain the maximum utility.Experiments results show that,our mechanism can achieve better performance in terms of incentive user participation and tasks coverage.2.In the above incentive mechanism design,some users may exit the sensing task randomly due to the randomness of smartphone users,which may result in low task completion ratio.Although using the above incentive mechanism to recruit more users to join in uncompleted sensing tasks will increase the task completion ratio,the cost of platform will be added.Thus,avoid increasing the cost of platform while improving the task completion ratio,a user-interaction based incentive model and a dynamic double auction framework are proposed.The simulation results show that the model can improve the task completion ratio for the MCS system.