Please use this identifier to cite or link to this item: https://doi.org/10.1109/ICCS.2010.5686092
Title: Access point placement for fingerprint-based localization
Authors: Sharma, C.
Wong, Y.F. 
Soh, W.-S. 
Wong, W.-C. 
Keywords: Access point
Indoor localization
Location fingerprint
Similar fingerprint
Simulated annealing
Issue Date: 2010
Source: Sharma, C.,Wong, Y.F.,Soh, W.-S.,Wong, W.-C. (2010). Access point placement for fingerprint-based localization. 12th IEEE International Conference on Communication Systems 2010, ICCS 2010 : 238-243. ScholarBank@NUS Repository. https://doi.org/10.1109/ICCS.2010.5686092
Abstract: Fingerprint-based localization systems mainly utilize the received signal strength (RSS) to estimate the location of a receiver, and the localization accuracy depends largely on the number and placement of access points (APs). In this paper, we propose a novel method for placing the APs, which aims to minimize the total number of similar fingerprints (SFs) over the entire array of training locations. Minimizing SFs will increase the diversity of RSS array and hence improve localization accuracy. To solve the problem using a simple brute force search would be highly computationally expensive and inefficient. Instead, we propose a heuristic optimization algorithm based on Simulated Annealing (SA). Numerical results are obtained for both the brute force search and the SA based approach. Finally, the proposed algorithm's outputs, i.e., the APs' locations, are used with a K-Nearest Neighbors based localization algorithm, and the resulting localization errors are analyzed. © 2010 IEEE.
Source Title: 12th IEEE International Conference on Communication Systems 2010, ICCS 2010
URI: http://scholarbank.nus.edu.sg/handle/10635/69140
ISBN: 9781424470068
DOI: 10.1109/ICCS.2010.5686092
Appears in Collections:Staff Publications

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

SCOPUSTM   
Citations

9
checked on Dec 5, 2017

Page view(s)

27
checked on Dec 9, 2017

Google ScholarTM

Check

Altmetric


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