As a promising paradigm, mobile crowdsensing exerts the potential of widespread sensors embedded in mobile devices. The greedy nature of workers brings the problem of low-quality sensing data, which poses threats to the overall performance of a crowdsensing system. Existing works often tackle this problem with additional function components. In this paper, we systematically formulate the problem into a crowdsensing interaction process between a requestor and a worker, which can be modeled by two types of iterated games with different strategy spaces. Considering that the low-quality data submitted by the workers can reduce the requestor's payoff and further decrease the global income, we turn to controlling the social welfare in the games. To that aim, we take advantage of zero-determinant strategy, based on which we propose two social welfare control mechanisms under both game models. Specifically, we consider the requestor as the controller of the games and, with proper parameter settings for the to-be-adopted zero-determinant strategy, social welfare can be optimized to the desired level no matter what strategy the worker adopts. Simulation results demonstrate that the requestor can achieve the maximized social welfare and keep it stable by using our proposed mechanisms.
