With the proliferation of sensor-rich smart devices (smartphones, ipads, etc.), Mobile Crowd Sensing (MCS) has gradually attracted much attention in the research community recently. Worker recruitment is a crucial research issue in MCS system, in which platform recruits workers and assigns sensing tasks to them. While previous studies focus on either opportunistic-sensing-based worker recruitment or participatory-sensing-based worker recruitment separately, we proposed a two-phase hybrid worker recruitment framework named HySelector, which recruits workers in two phases. First, in the offline phase, borrowing the idea of influence propagation in communication and social network, we proposed algorithm to recruit opportunistic workers during their daily routines which can alleviate the cold start problem in traditional MCS system. Then, in the online phase, in order to reduce the computational complexity, we devised algorithm to incentivize participatory workers to move to specific subareas obtained by subareas clustering to fulfil sensing tasks. In both phases, we considered guaranteeing the incentive cost and time constraint. Experimental results on two open datasets demonstrated that compared with other methods, HySelector had better performance in terms of spatial coverage and running time under budget constraints.

随着传感器丰富的智能设备（智能手机，ipad等）的激增，移动人群感应（MCS）逐渐引起了研究界的广泛关注。工人招募是MCS系统中的关键研究问题，在该系统中，平台招募工人并向他们分配感知任务。尽管先前的研究分别关注基于机会感知的工人招募或基于参与感知的工人招募，但我们提出了一个名为HySelector的两阶段混合型工人招募框架，该框架分两个阶段招募工人。首先，在离线阶段，借鉴沟通和社交网络中影响力传播的思想，提出了一种在日常工作中招募机会性工人的算法，可以缓解传统MCS系统的冷启动问题。然后，在在线阶段，为了降低计算复杂度，我们设计了一种算法来激励参与工作的人员迁移到通过子区域聚类获得的特定子区域来完成传感任务。在两个阶段中，我们都考虑了保证激励成本和时间约束。在两个开放数据集上的实验结果表明，与其他方法相比，HySelector在预算约束下的空间覆盖范围和运行时间方面具有更好的性能。