期刊名称：Lecture Notes in Engineering and Computer Science
印刷版ISSN：2078-0958
电子版ISSN：2078-0966
出版年度：2018
卷号：2235&2236
页码：380-384
出版社：Newswood and International Association of Engineers
摘要：In recent years, most of the existing Indoor Localization
Systems (ILSs) mainly used the Wi-Fi based fingerprinting
approach to determine the floor position in multi-floor
buildings. However, this method suffers from large fluctuation
on the Wi-Fi signals which reduces the accuracy of floor
estimation. Besides, it is also computationally extensive due to
its heavy database size. For resolving these issues, we propose an
improved floor determination technique using the atmospheric
pressure sensor built in a smartphone. In this paper, the
statistical features are extracted from the atmospheric pressure
data, then the Fuzzy C-Mean (FCM) clustering algorithm is
applied on them to improve the floor determination technique.
The proposed work has been evaluated on the “Indoor Positioning
Indoor Navigation” (IPIN) 2016 competition dataset.
Experimental results are compared with three other well known
methods and the participants of IPIN 2016. The proposed
method exhibited better results for floor estimation with the
accuracy of 98.68%, and the time complexity of proposed
method also improved to O(1).
关键词：Location tracking; Received Signal Strength
(RSS); Radio map; Wi;Fi; Mobile device;
