基于移动群体感知的参与者选择算法

发布时间：2018-06-05 17:01

本文选题：移动群体感知 + 激励机制　；参考：《河北大学》2017年硕士论文

【摘要】：移动群体感知是移动计算与移动互联网领域的研究热点,其基本思想是利用智能终端收集周围环境的感知数据,并将这些感知数据应用于各种创新性应用。感知信息质量是决定移动群体感知性能的关键,而其中参与者选择算法和激励约束是影响感知信息质量的两个重要因素,参与者选择策略的目标是选择有限数量的最优移动节点,采集覆盖最全面的感知数据。从这个目标出发,本文引入基于加权熵的多目标决策理论,设计基于加权熵多目标决策理论的参与者选择算法,对感知节点的感知能力、转移到目标感知区域的转移概率和激励要求进行综合评定,从而选择出最优质的感知节点来完成多感知任务。首先,构建了移动群体感知系统模型,该系统模型主要由发布感知任务的个人或实体、移动群体感知数据服务中心和感知节点组成。其次,根据连续马尔科夫链原理建立感知节点的移动模型。最后,引入基于加权熵的多目标决策理论,设计了基于加权熵多目标决策理论的参与者选择算法MODSM,为了验证算法的性能,本文应用Python脚本语言进行实验仿真,验证了所设计方法的有效性。
[Abstract]:Mobile group awareness is a hot topic in the field of mobile computing and mobile Internet. Its basic idea is to use intelligent terminals to collect the perceptual data of the surrounding environment and apply these perceptual data to various innovative applications. Perceived information quality is the key to determine the perceived performance of mobile groups, and participant selection algorithm and incentive constraints are two important factors that affect the perceived information quality. The goal of the participant selection strategy is to select a limited number of optimal mobile nodes and collect the most comprehensive perceptual data. From this goal, this paper introduces the multi-objective decision-making theory based on weighted entropy, designs the participant selection algorithm based on the weighted entropy multi-objective decision making theory, and the perception ability of the perceived nodes. The transition probability and incentive requirement of moving to the target perceptual region are evaluated synthetically to select the best perception node to complete the multi-perception task. First of all, a mobile group awareness system model is constructed. The system model is composed of the individual or entity that issues the perceptual task, the mobile group perception data service center and the perception node. Secondly, the moving model of perceptual nodes is established according to the principle of continuous Markov chain. Finally, the multi-objective decision-making theory based on weighted entropy is introduced, and the participant selection algorithm based on weighted entropy multi-objective decision theory is designed. In order to verify the performance of the algorithm, this paper uses Python script language to carry out experimental simulation. The validity of the proposed method is verified.
【学位授予单位】：河北大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP301.6

【参考文献】