基于用户分类可靠性的群智感知任务分配机制研究  

An allocation mechanism based on the classification reliability of users for crowdsensing systems

作　　者：卜霄菲[1,2] 吴文江 辛煜[3] 黄河[4,5] 吴晓灿[4] 杨雪华 

机构地区：[1]中国科学院沈阳计算技术研究所有限公司,辽宁沈阳110168 [2]沈阳师范大学软件学院,辽宁沈阳110034 [3]北京遥感信息研究所,北京100011 [4]苏州大学计算机科学与技术学院,江苏苏州215006 [5]中国科学技术大学苏州研究院,江苏苏州215123

出　　处：《安徽大学学报（自然科学版）》2018年第2期67-73,共7页Journal of Anhui University(Natural Science Edition)

基　　金：辽宁省高等学校基本科研专项资金资助项目(LQN201707)

摘　　要：群智感知系统利用智能移动终端上的传感器,可以快速、便捷、低成本地收集到大量的数据.任务分配是群智感知中的关键技术,然而现有研究对用户的可靠性问题考虑不足,所设计出的分配机制难以满足实际系统的需求.针对该问题,文章假设用户所提交的数据可能是不可靠的,且同一个用户完成不同类型任务的可靠性也有差异.所设计的机制迭代进行,每次迭代以实现数据消费者的最大最小公平作为优化目标,贪心地选择一个任务与用户完成匹配,直到所有数据消费者的收益均无法提高为止.在每次任务完成后,所设计机制还会根据用户所提交的数据质量不断更新用户的分类可靠性.最后,通过仿真实验验证了所设计机制的有效性.Crowdsensing system uses the sensors on intelligent mobile terminals and it can collect a large amount of data quickly, conveniently and cheaply. Task allocation is the key technology of crowdsensing. However, the existing studies have not fully considered user＇s reliability. Thus, the designed allocation mechanisms cannot meet the needs of the actual system. In this paper, we assumed that the data submitted by users may be unreliable, and the reliability of the same user to complete different types of tasks is also different. The designed mechanism runs periodically. In each iteration, the designed mechanism greedily selects a task and a user, which can maximize the minimum utility of data consumers, until all the utilities of data consumers are unable to improve so far. After all the tasks are completed, the reliabilities of users will be updated according to the quality of the data submitted by them, Finally, the simulation results verified the effectiveness of the proposed mechanism.

关 键 词：任务分配 群智感知 可靠性 最大最小公平 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]