A Group Buying Based Incentive Mechanism For Mobile Crowd Sensing

With the growing popularity of portable smart devices,mobile crowd sensing(MCS)has become a new paradigm to perceive the real physical world.Leveraging human intelligence and built-in sensors of smart devices,MCS is able to collect sensing data related to people and extract valuable information from the data.The information can be helpful for environmental monitoring,transportation planning and so on.With the help of MCS,our living quality can be considerably improved.In the MCS system,the sensing data are contributed by ordinary users.And it takes effort to collect the required sensing data.Thus the MCS system needs to provide incentives to encourage participants.The monetary award is an effective incentive mechanism for stimulating participation in MCS.However,the data collection for a long period of time and a wide area requires large monetary awards,which would be unaffordable for data requesters.We propose a Two-phase Group Buying based Auction(TGBA)mechanism.The data requesters pay for the sensing data in a group buying way,which would reduce their payments.The similar requests form a group and cooperate with each other to hire sensing users and share the sensing data and the cost.In phase I of TGBA,the auctions are held in each group independently.The group agent needs to decide the winners of the auction and the payments for winners.In phase II,the group agents bid for sensing users and the group buying platform should decide the allocation of sensing users and the payments of group agents.For the auction in Phase I,we propose Random Single Unit Clearing Price(RSUCP)algorithm and Random Sampling Profit Extraction(RSPE)algorithm.RSUCP brings higher utility for group members and group agents,while RSPE brings higher utility for sensing users and higher number of successful deals.For the auction in Phase II,we propose Random Assignment and Differential Pricing(RADP)algorithm and Random Assignment and K-th Differential Pricing(RAKDP)algorithm.RAKDP achieves higher utility for sensing users.TGBA achieves individual rationality,budget balance and truthfulness and is computational efficient.Simulation results shows that TGBA is a win-win mechanism for data requesters and improves the utility of sensing users.TGBA has good performance in the scenario of MCS and can be applied in the practical MCS systems.