| 引用本文: | 蒋伟进,吕斯健,陈晓红.群智感知中面向移动群体的参与者选择优化模型[J].控制理论与应用,2022,39(2):343~351.[点击复制] |
| JIANG Wei-jin,LV Si-jian,CHEN Xiao-hong.Participant selection optimization model for mobile groups in crowdsensing[J].Control Theory and Technology,2022,39(2):343~351.[点击复制] |
|
| 群智感知中面向移动群体的参与者选择优化模型 |
| Participant selection optimization model for mobile groups in crowdsensing |
| 摘要点击 1396 全文点击 439 投稿时间：2020-09-07 修订日期：2022-01-24 |
| 查看全文 查看/发表评论 下载PDF阅读器 |
| DOI编号 10.7641/CTA.2021.00600 |
| 2022,39(2):343-351 |
| 中文关键词 数据驱动 群智感知 密度聚类 优化算法 |
| 英文关键词 data-driven crowdsensing density clustering optimization |
| 基金项目 国家自然科学基金面上项目(61772196, 61472136), 湖南省自然科学基金重点项目(2020JJ4249), 湖南省社会科学基金重点项目(2016ZDB006), 湖 南省社会科学成果评审委员会课题重点项目(湘社评19ZD1005), 湖南省学位与研究生教育改革研究基金资助项目(2020JGYB234)资助. |
|
| 中文摘要 |
| 随着短视频时代的来临, 移动群智感知任务的视频化程度越来越高, 在传统研究中常利用机会网络和移动
网络激励任务的分发和数据的收集, 但机会网络中节点移动的不可控性, 以及视频任务内容传输的高代价性都使得
这些方法的实用性大大降低. 针对此问题, 利用社会移动群体规律性的自主聚集、活动范围大等特点, 提出一种面
向社会移动群体的群智感知参与者选择优化模型. 利用密度聚类算法根据同类任务的位置进行划分得出聚类中心,
实现任务子区域所属地铁站点的划分. 包括基于用户激励成本的参与者优化算法和基于用户数量的参与者优化算
法. 仿真结果表明, 与同类算法相比可以消耗更低的系统资源选择出参与者数量更少的任务分发方案. |
| 英文摘要 |
| With the advent of short video era, mobile group awareness tasks are becoming more and more videointensive.
Opportunity networks and mobile networks are often used in traditional research to motivate task distribution
and data collection. However, the uncontrollability of node movement in the opportunity network and the high cost of
video task content transmission make these methods less practical. In order to solve this problem, an optimization model
of crowdsensing perception participant selection for social mobile groups is presented, which takes advantage of the regularity
of autonomous clustering and large range of activities of social mobile groups. The density clustering algorithm is
used to divide the cluster centers according to the locations of similar tasks, so as to divide the metro stations belonging
to the task subarea. It includes an optimization algorithm for participants based on user incentive cost and an optimization
algorithm for participants based on number of users. The simulation results show that a task distribution scheme with fewer
participants can be selected using less system resources than similar algorithms. |
|
|
|
|
|