• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2015, Vol. 37 ›› Issue (01): 184-190.

• 论文 • 上一篇    

物联网环境下基于用户满意度的实时调度算法

王盖,王永炎   

  1. (中国科学院软件研究所,北京 100190)
  • 收稿日期:2014-07-30 修回日期:2014-09-14 出版日期:2015-01-25 发布日期:2015-01-25
  • 基金资助:

    国家自然科学基金面上资助项目(61273080);国家科技重大项目(2012ZX01039004)

Real-time scheduling algorithm for
Internet of Things based on user satisfaction  

WANG Gai,WANG Yongyan   

  1. (Institute of Software,Chinese Academy of Sciences,Beijing 100190,China)
  • Received:2014-07-30 Revised:2014-09-14 Online:2015-01-25 Published:2015-01-25

摘要:

随着3G网络和移动技术的发展,物联网已经深入到人们的工作和生活中,提供无所不在的服务。用户作为服务对象,以追求满意的高品质服务为目标,对服务具有选择权,对服务品质具有最直接、最重要的评判权。系统作为服务提供者,其服务品质取决于每个用户的满意度。因此,如何通过调度算法合理分配系统资源,让每个用户获得满意度更高的服务,成为了研究的重点。首先,构建个人满意度和系统满意度的评价标准;其次,依据用户期望的服务水平,为任务分配优先级;最后,通过反馈调节机制来调整任务优先级的分配,让更多用户对服务满意。实验表明,物联网环境下基于用户满意度的实时调度算法能够根据用户的需求,为用户提供满意的差别化服务,并提高系统满意度。

关键词: 用户满意度, 服务质量, 物联网, 反馈调节, 模糊综合评判, 实时调度

Abstract:

With the development of 3G networks and mobile telecommunications technology, Internet of Things (IoT) has penetrated into people’s life providing ubiquitous service. As the center of service, users always pursue service of high quality and they have the most direct and powerful right to evaluate a service. As service providers, computer systems’ service quality is based on the level of every user’s satisfaction. Therefore, it is crucial to construct a scheduling algorithm for allocating system resources reasonably to guarantee a better service. Firstly, we establish the quantitative evaluation standards for personal satisfaction and system satisfaction. Secondly, we assign priorities for tasks according to the service quality that users expect. Finally, we use feedback regulations to adjust task priority assignment, so that more users will be satisfied with the provided service. Experimental results show that the proposed algorithm can help computer systems offer differentiated services based on users’ needs and ensures a higher system satisfaction.

Key words: user satisfaction;service quality;Internet of Things;feedback regulation;fuzzy comprehensive evaluation method;realtime schedule