检索结果-内蒙古大学图书馆

Dirichlet tessellation for high-dimensional RSS clustering in Wi-Fi environment

ICIC Express Letters 2014年第7期8卷 2043-2048页

作者： Zhou, Mu Tian, Zengshan Xu, Kunjie Tang, Xiaomou Chongqing Key Lab of Mobile Communications Technology Chongqing University of Posts and Telecommunications No. 2 Chongwen Road Nanan District Chongqing 400065 China Department of Electronic and Computer Engineering The Hong Kong University of Science and Technology Clear Water Bay Kowloon Hong Kong Graduate Telecommunication and Networking Program School of Information Science University of Pittsburgh PA 15260 United States

The received signal strength (RSS) clustering has played a significant role in site surveying and data pre-processing for the location tracking in Wi-Fi environment. To this end, this paper presents a novel clustering algorithm to cluster the raw RSS vectors with large similarities in both the RSS and physical locations together. The clusters of RSS vectors are determined by the cells which are generated by the Dirichlet tessellation in a transformed low-dimensional space. Our algorithm consists of two steps. We first use the Laplacian embedding to transform the raw measurements into a low-dimensional space, while preserving the RSS and location neighborhood relations. Second, after the seeds of cells are carefully selected, we will tessellate the low-dimensional space by the Dirichlet cells to obtain the clustering results. The experiments conducted in an actual public Wi-Fi environment prove that our algorithm can achieve the smallest cluster scattering in both the RSS and locations compared with the existing affinity propagation clustering, K-means clustering and binary support vector classifier (SVC) algorithms. © 2014 ISSN 1881-803X.

关键词： Wi-Fi

来源：评论

学校读者我要写书评

暂无评论

Towards statistical linear accuracy of neighbor matching localization in Wi-Fi environment

引用

Journal of Computational Information Systems 2013年第22期9卷 9177-9184页

作者： Zhou, Mu Tian, Zengshan Xu, Kunjie Hong, Xia Tang, Xiaomou Chongqing Key Lab. of Mobile Communications Technology School of Communication and Information Engineering Chongqing University of Posts and Telecommunications Chongqing 400065 China Graduate Telecommunication and Networking Program School of Information Science University of Pittsburgh Pittsburgh 15260 United States

As one of the most significant work of interest for the pervasive computing and ubiquitous context-awareness, the ways to effectively locate people's positions and realize the real-time interaction among different location-based services and devices have always been seriously concerned. Although a large number of localization systems for outdoor environments have been proposed in the past decade ranging from the military to public uses, their accuracy in indoor and urban environments can not be guaranteed because of the received signal strength (RSS) shadowing and multi-path interference. Therefore, the neighbor matching localization using Wi-Fi network in indoor areas has been significantly favored since the invention of the world's first RSS fingerprints-based RADAR localization system in 2000. However, there is no rigorous analytical solution to the statistical errors of neighbor matching localization. Therefore, we will present some mathematical analysis towards the statistical linear errors of neighbor matching localization and introduce the position entropy as a new metric to examine the variations of accuracy. Furthermore, the mathematical results and some related interesting observations from this paper are also verified by the numerical and simulation evaluations. © 2013 Binary Information Press.

关键词： Entropy

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：