移动群智感知中基于任务质量的多任务分发参与者选择

任务分发作为移动群智感知领域的关键研究问题是目前的热点和难点,针对该领域中多任务分发条件下的参与者选择方法进行研究,根据参与者的历史任务完成情况,利用参与者累计信誉以及参与意愿构建参与者服务质量模型(QoS).在最大化QoS的基础上,采用启发式贪心算法解决参与者的选择问题.提出以任务为中心和以用户为中心的两种选择方案,在算法中引入距离约束因子、完整度约束因子以及信誉度约束因子,在保证任务完成质量的前提下,尽可能提升平台最终收益和参与者的收益.对两种算法的可行性和有效性进行评估,与现有的算法在选择出的参与者人数、移动距离以及数据质量等方面进行详细对比,分析各种因素对选择结果的影响.在实验过程中,建立阶跃型数据定价模型,对参与者上传的数据质量进行量化对比.实验结果表明,所提出的算法较现有的算法在服务质量方面取得了很好的效果.

Multitask-oriented participant selection based on task quality in mobile crowd sensing

As a key research issue, task distribution is a hot and difficult point of current research in the field of mobile crowding sensing(MCS). The method of multi-task distribution participant selection is studied in this paper. According to the historical task completion status of participants, the cumulative reputation and willingness of participants are used to build the quality of service model(QoS). On the basis of maximizing the QoS, the greedy heuristic algorithm is used to solve the participation problem, and the two design schemes including task-centric and user-centric are proposed. The distance constraint factor, integrity constraint factor and reputation constraint factor are introduced into the proposed algorithms. The purpose is to improve the quality of perception tasks as much as possible, so as to improve the final benefits of platforms. Two algorithms are evaluated in the feasibility and effectiveness, and compared with existing algorithms in terms of the number of participants selected, moving distance, and data quality. During the experiment, a step data pricing model is established to quantitatively compare the quality of data uploaded by participants. Experimental results show that the two algorithms proposed are better than the existing algorithms in task quality.