Incentive Mechanism Design for Edge-Cloud Collaboration in Mobile Crowd Sensing

In this paper, we propose a Pricing-based Incentive Mechanism for Edge-Cloud collaboration (PIM-EC) in mobile crowd sensing. In PIM-EC, data can be exchanged among different regions with the support of edge-cloud servers, which improves the data efficiency. For the utility conflicts between mobile sensing users and cloud servers in the traditional one-stage game, we design a two-stage game which includes the bargaining game between users and edge-cloud servers, as well as a data trading game among different edge-cloud servers deployed in different regions. For the first stage game, based on Stackelberg game model and Rubinstein bargaining model, we design a finite-period dynamic bargaining algorithm. For the second stage game, based on the optimal auction mechanism and using the augmented Lagrange multiplier method, a quasi-Newton iterative pricing algorithm is proposed. We investigate the performance of PIM-EC through simulations. Compared with SWMA and IMC-SS, the social welfare of PIM-EC is increased by 21% and 34% respectively.