An accuracy enhancement algorithm for fingerprinting method

Yuntian Brian Bai; Mani Williams; Falin Wu; Allison Kealy; Kefei Zhang

doi:10.1109/DSAA.2014.7058060

An accuracy enhancement algorithm for fingerprinting method

Yuntian Brian Bai^*
, Mani Williams
, Falin Wu
, Allison Kealy
, Kefei Zhang

^*Corresponding author for this work

School of Instrumentation and Optoelectronic Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Scopus citations

Abstract

Fingerprinting is the prevailing positioning method for location based service (LBS) and indoor positioning applications when compared with other methods such as cell of origin (CoO) and trilateration. It is especially more suitable for complicated indoor environments. However, higher positioning accuracy is still expected for it to match the capabilities of other mature techniques such as GPS. This paper presents a new algorithm for improving the positioning accuracy of the Nearest Neighbour (NN) algorithm from a Wi-Fi-based fingerprinting method. The new algorithm initially used the NN algorithm to identify the initial position estimate of the user being tracked. Then two distance correction values in two roughly perpendicular directions were calculated by the path loss model based on the two signal strength indicator (RSSI) values observed. The errors from the path loss model were eliminated through differencing two calculated distances which were derived from a similar environment. The new algorithm was tested and the results evaluated against that of the NN algorithm. The preliminary results from 24 test points showed that the positioning accuracy of the new approach has improved consistently and the root mean square accuracy improved to 3.4 m from 3.8 m with the NN algorithm.

Original language	English
Title of host publication	DSAA 2014 - Proceedings of the 2014 IEEE International Conference on Data Science and Advanced Analytics
Editors	Longbing Cao, George Karypis, Irwin King, Wei Wang
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	110-114
Number of pages	5
ISBN (Electronic)	9781479969913
DOIs	https://doi.org/10.1109/DSAA.2014.7058060
State	Published - 10 Mar 2014
Event	1st IEEE International Conference on Data Science and Advanced Analytics, DSAA 2014 - Shanghai, China Duration: 30 Oct 2014 → 1 Nov 2014

Publication series

Name	DSAA 2014 - Proceedings of the 2014 IEEE International Conference on Data Science and Advanced Analytics

Conference

Conference	1st IEEE International Conference on Data Science and Advanced Analytics, DSAA 2014
Country/Territory	China
City	Shanghai
Period	30/10/14 → 1/11/14

Keywords

Fingerprinting
Indoor positioning
LBS
Wi-Fi
positioning algorithm

Access to Document

10.1109/DSAA.2014.7058060

Cite this

Bai, Y. B., Williams, M., Wu, F., Kealy, A., & Zhang, K. (2014). An accuracy enhancement algorithm for fingerprinting method. In L. Cao, G. Karypis, I. King, & W. Wang (Eds.), DSAA 2014 - Proceedings of the 2014 IEEE International Conference on Data Science and Advanced Analytics (pp. 110-114). Article 7058060 (DSAA 2014 - Proceedings of the 2014 IEEE International Conference on Data Science and Advanced Analytics). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/DSAA.2014.7058060

Bai, Yuntian Brian ; Williams, Mani ; Wu, Falin et al. / An accuracy enhancement algorithm for fingerprinting method. DSAA 2014 - Proceedings of the 2014 IEEE International Conference on Data Science and Advanced Analytics. editor / Longbing Cao ; George Karypis ; Irwin King ; Wei Wang. Institute of Electrical and Electronics Engineers Inc., 2014. pp. 110-114 (DSAA 2014 - Proceedings of the 2014 IEEE International Conference on Data Science and Advanced Analytics).

@inproceedings{b51fd67a4e864b9885be2954ff2549e3,

title = "An accuracy enhancement algorithm for fingerprinting method",

abstract = "Fingerprinting is the prevailing positioning method for location based service (LBS) and indoor positioning applications when compared with other methods such as cell of origin (CoO) and trilateration. It is especially more suitable for complicated indoor environments. However, higher positioning accuracy is still expected for it to match the capabilities of other mature techniques such as GPS. This paper presents a new algorithm for improving the positioning accuracy of the Nearest Neighbour (NN) algorithm from a Wi-Fi-based fingerprinting method. The new algorithm initially used the NN algorithm to identify the initial position estimate of the user being tracked. Then two distance correction values in two roughly perpendicular directions were calculated by the path loss model based on the two signal strength indicator (RSSI) values observed. The errors from the path loss model were eliminated through differencing two calculated distances which were derived from a similar environment. The new algorithm was tested and the results evaluated against that of the NN algorithm. The preliminary results from 24 test points showed that the positioning accuracy of the new approach has improved consistently and the root mean square accuracy improved to 3.4 m from 3.8 m with the NN algorithm.",

keywords = "Fingerprinting, Indoor positioning, LBS, Wi-Fi, positioning algorithm",

author = "Bai, \{Yuntian Brian\} and Mani Williams and Falin Wu and Allison Kealy and Kefei Zhang",

note = "Publisher Copyright: {\textcopyright} 2014 IEEE.; 1st IEEE International Conference on Data Science and Advanced Analytics, DSAA 2014 ; Conference date: 30-10-2014 Through 01-11-2014",

year = "2014",

month = mar,

day = "10",

doi = "10.1109/DSAA.2014.7058060",

language = "英语",

series = "DSAA 2014 - Proceedings of the 2014 IEEE International Conference on Data Science and Advanced Analytics",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "110--114",

editor = "Longbing Cao and George Karypis and Irwin King and Wei Wang",

booktitle = "DSAA 2014 - Proceedings of the 2014 IEEE International Conference on Data Science and Advanced Analytics",

address = "美国",

}

Bai, YB, Williams, M, Wu, F, Kealy, A & Zhang, K 2014, An accuracy enhancement algorithm for fingerprinting method. in L Cao, G Karypis, I King & W Wang (eds), DSAA 2014 - Proceedings of the 2014 IEEE International Conference on Data Science and Advanced Analytics., 7058060, DSAA 2014 - Proceedings of the 2014 IEEE International Conference on Data Science and Advanced Analytics, Institute of Electrical and Electronics Engineers Inc., pp. 110-114, 1st IEEE International Conference on Data Science and Advanced Analytics, DSAA 2014, Shanghai, China, 30/10/14. https://doi.org/10.1109/DSAA.2014.7058060

An accuracy enhancement algorithm for fingerprinting method. / Bai, Yuntian Brian; Williams, Mani; Wu, Falin et al.
DSAA 2014 - Proceedings of the 2014 IEEE International Conference on Data Science and Advanced Analytics. ed. / Longbing Cao; George Karypis; Irwin King; Wei Wang. Institute of Electrical and Electronics Engineers Inc., 2014. p. 110-114 7058060 (DSAA 2014 - Proceedings of the 2014 IEEE International Conference on Data Science and Advanced Analytics).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - An accuracy enhancement algorithm for fingerprinting method

AU - Bai, Yuntian Brian

AU - Williams, Mani

AU - Wu, Falin

AU - Kealy, Allison

AU - Zhang, Kefei

PY - 2014/3/10

Y1 - 2014/3/10

N2 - Fingerprinting is the prevailing positioning method for location based service (LBS) and indoor positioning applications when compared with other methods such as cell of origin (CoO) and trilateration. It is especially more suitable for complicated indoor environments. However, higher positioning accuracy is still expected for it to match the capabilities of other mature techniques such as GPS. This paper presents a new algorithm for improving the positioning accuracy of the Nearest Neighbour (NN) algorithm from a Wi-Fi-based fingerprinting method. The new algorithm initially used the NN algorithm to identify the initial position estimate of the user being tracked. Then two distance correction values in two roughly perpendicular directions were calculated by the path loss model based on the two signal strength indicator (RSSI) values observed. The errors from the path loss model were eliminated through differencing two calculated distances which were derived from a similar environment. The new algorithm was tested and the results evaluated against that of the NN algorithm. The preliminary results from 24 test points showed that the positioning accuracy of the new approach has improved consistently and the root mean square accuracy improved to 3.4 m from 3.8 m with the NN algorithm.

AB - Fingerprinting is the prevailing positioning method for location based service (LBS) and indoor positioning applications when compared with other methods such as cell of origin (CoO) and trilateration. It is especially more suitable for complicated indoor environments. However, higher positioning accuracy is still expected for it to match the capabilities of other mature techniques such as GPS. This paper presents a new algorithm for improving the positioning accuracy of the Nearest Neighbour (NN) algorithm from a Wi-Fi-based fingerprinting method. The new algorithm initially used the NN algorithm to identify the initial position estimate of the user being tracked. Then two distance correction values in two roughly perpendicular directions were calculated by the path loss model based on the two signal strength indicator (RSSI) values observed. The errors from the path loss model were eliminated through differencing two calculated distances which were derived from a similar environment. The new algorithm was tested and the results evaluated against that of the NN algorithm. The preliminary results from 24 test points showed that the positioning accuracy of the new approach has improved consistently and the root mean square accuracy improved to 3.4 m from 3.8 m with the NN algorithm.

KW - Fingerprinting

KW - Indoor positioning

KW - LBS

KW - Wi-Fi

KW - positioning algorithm

UR - https://www.scopus.com/pages/publications/84946692390

U2 - 10.1109/DSAA.2014.7058060

DO - 10.1109/DSAA.2014.7058060

M3 - 会议稿件

AN - SCOPUS:84946692390

T3 - DSAA 2014 - Proceedings of the 2014 IEEE International Conference on Data Science and Advanced Analytics

SP - 110

EP - 114

BT - DSAA 2014 - Proceedings of the 2014 IEEE International Conference on Data Science and Advanced Analytics

A2 - Cao, Longbing

A2 - Karypis, George

A2 - King, Irwin

A2 - Wang, Wei

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 1st IEEE International Conference on Data Science and Advanced Analytics, DSAA 2014

Y2 - 30 October 2014 through 1 November 2014

ER -

Bai YB, Williams M, Wu F, Kealy A, Zhang K. An accuracy enhancement algorithm for fingerprinting method. In Cao L, Karypis G, King I, Wang W, editors, DSAA 2014 - Proceedings of the 2014 IEEE International Conference on Data Science and Advanced Analytics. Institute of Electrical and Electronics Engineers Inc. 2014. p. 110-114. 7058060. (DSAA 2014 - Proceedings of the 2014 IEEE International Conference on Data Science and Advanced Analytics). doi: 10.1109/DSAA.2014.7058060

An accuracy enhancement algorithm for fingerprinting method

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this