Boosting task completion rate for time-sensitive MCS system

In Mobile Crowdsensing system, many sensing tasks are time-sensitive, hence, time validity is essential cause data outside the time frame is useless which will not only increase requesters' costs but also damage requesters' efficiency and utilities. Therefore, it is crucial to design an incentive mechanism to increase system's task completion rate within time limitations. A dilemma arises between requesters and the MCS system due to the trad-off between requester's goal to maximize its own utilities and the MCS system's purpose to increase the whole system's task completion rate. To solve the dilemma, we come up with a dynamic coopetition (competition and cooperation) incentive mechanism where in every stage there are competition to assure requesters' utilities and cooperation to reallocate data to advance the system's task completion rate. During competition, a matching algorithm with budget constraints is presented to derive the optimal matching strategies for requesters and participants without exceeding requesters' budgets while maximizing utilities. After competing, the centralized MCS platform would reallocate requesters' excessive data to those who failed to get enough data on time through a data reallocation algorithm. By simulations, we compare the performance of the coopetition incentive mechanism with different orders and parameters. Regardless of parameters or orders, our algorithm can promise a minimum task completion rate of 94% compared with less than 32% task completion rate with only competition. Among all three orders, emergency degree has the best performance with task completion rate up to 99%.