Journal
INTERNATIONAL JOURNAL OF FUTURE GENERATION COMMUNICATION AND NETWORKING
Volume 10, Issue 3, Pages 1-9Publisher
NADIA
DOI: 10.14257/ijfgcn.2017.10.3.01
Keywords
Indoor Positioning; FCM; Whitening method; accuracy; computation complexity
Categories
Funding
- Science and Technology Innovation Fund Qingdao Science and Technology Development Project [KJZD-13-29-JCH]
- Key Science and Technology Project of Qingdao Economic and Technological Development Zone [2013-1-25]
- Nature Science Foundation of China Province [71403151]
- Nature Science Foundation of Shandong Province [ZR2013FM023]
- Technology Innovation Fund Project of Shandong University of Science and Technology [YC150333]
Ask authors/readers for more resources
Due to pervasive penetration of the Wireless Local Area Networks (WLAN) and portable smartphones, the indoor positioning technology based on WLAN has developed rapidly. Especially the Wi-Fi fingerprint-based positioning has attracted continuous attention because of its low cost, high applicability and accuracy. The current fingerprint-based positioning algorithm still has a tremendous challenge both in accuracy and efficiency. To improve and optimize the issues, a novel indoor positioning algorithm based on Whitening RSS and Fuzzy C-Means (FCM) clustering approach is proposed in this paper. Firstly, in offline phase, the preprocessing of gathered RSS data samples is conducted by whitening methods, removing the correlation of the signals, to build the fingerprint database. Then the computation complexity is reduced by dividing the whole database into various sub-databases through FCM clustering algorithm. Meanwhile, the membership degrees obtained from FCM are added to the position calculation as a weight parameter to improve accuracy. Simulation results indicate that the proposed algorithm extraordinary achieves high performance in reducing the computation complexity and enhancing the positioning accuracy.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available