Please use this identifier to cite or link to this item: https://doi.org/10.1109/CyberC.2013.63
Title: Improved PCA based step direction estimation for dead-reckoning localization
Authors: Bao, H.
Wong, W.-C. 
Keywords: Adaptive method
PCA
Sensor's orientation analysis
Step counting localization
Issue Date: 2013
Source: Bao, H., Wong, W.-C. (2013). Improved PCA based step direction estimation for dead-reckoning localization. Proceedings - 2013 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery, CyberC 2013 : 325-331. ScholarBank@NUS Repository. https://doi.org/10.1109/CyberC.2013.63
Abstract: Step direction estimation is one of the key procedures for step counting based dead-reckoning tracking using motion sensors. It is also quite challenging, especially when the captured motion data is tainted by the user's activity. The Principal Component Analysis (PCA) based algorithm has provided robust estimation results, regardless of the sensor's relative rotation compared to the human body. However, the PCA based algorithm only returns the principal axis, resolving the 180° ambiguity is another challenge. In this paper, the drawback of PCA is compensated with the sensor's orientation analysis, which returns the walking direction by analyzing the change of the sensor's orientation. In our adaptive method, the sensor's orientation analysis algorithm is executed when a direction change is detected by the PCA algorithm. Because of the low computational complexity and restricted usage of orientation analysis, the adaptive method introduces little overhead compared to the original PCA method. Experimental results show that the adaptive algorithm provides more robust and accurate results compared to the PCA algorithm. © 2013 IEEE.
Source Title: Proceedings - 2013 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery, CyberC 2013
URI: http://scholarbank.nus.edu.sg/handle/10635/83829
DOI: 10.1109/CyberC.2013.63
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
There are no files associated with this item.

SCOPUSTM   
Citations

3
checked on Feb 15, 2018

WEB OF SCIENCETM
Citations

2
checked on Jan 30, 2018

Page view(s)

18
checked on Feb 19, 2018

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.