Accurate WiFi localization by unsupervised fusion of extended candidate location set

Xiansheng Guo, Shilin Zhu, Lin Li, Fangzi Hu, Nirwan Ansari

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Fusing the predictions of multiple received signal strength (RSS)-based classifiers is an efficient strategy to mitigate the impact of the fluctuation of the RSS. However, most of the existing fusion methods exhibit two remarkable shortcomings: 1) they need to train and store offline weights by the supervised learning and 2) they directly fuse the so-called candidate location set (CLS), which is collected from the most likely location estimate of each classifier (location with the largest probability of being the true location predicted by the classifier), and thus do not fully leverage the knowledge of classifiers. In general, the fluctuation of RSS does not guarantee the location predicted by each classifier with the highest probability to be the true location, thus leading to severe performance degeneration of the existing fusion methods. To overcome the above shortcomings, we propose an accurate WiFi localization framework by unsupervised fusion of an extended CLS (ECLS). First, we train multiple classifiers by only using RSS fingerprints in the offline phase. In the online phase, instead of collecting the CLS from the trained classifiers, we construct an ECLS by augmenting CLS with other location estimates (locations with predication probability greater than a certain threshold) from each classifier. As compared with the CLS, ECLS provides a bigger fusion space that likely includes the true location of the user. Furthermore, an unsupervised fusion localization algorithm based on the ECLS is derived from the joint optimization of weights and the location of the user. Furthermore, a point of inflection searching algorithm is also proposed to intelligently construct the ECLS. Real experimental results show that our proposed algorithm is more robust to changing environments and model errors, and can significantly improve the localization accuracy without any fingerprint and hardware calibrations.

Original languageEnglish (US)
Article number8466583
Pages (from-to)2476-2485
Number of pages10
JournalIEEE Internet of Things Journal
Volume6
Issue number2
DOIs
StatePublished - Apr 1 2019

Fingerprint

Fusion reactions
Classifiers
Supervised learning
Electric fuses
Calibration
Hardware

All Science Journal Classification (ASJC) codes

  • Information Systems
  • Signal Processing
  • Hardware and Architecture
  • Computer Networks and Communications
  • Computer Science Applications

Keywords

  • Candidate location set (CLS)
  • WiFi
  • extended candidate location set (ECLS)
  • indoor localization
  • received signal strength (RSS)
  • unsupervised fusion

Cite this

Guo, Xiansheng ; Zhu, Shilin ; Li, Lin ; Hu, Fangzi ; Ansari, Nirwan. / Accurate WiFi localization by unsupervised fusion of extended candidate location set. In: IEEE Internet of Things Journal. 2019 ; Vol. 6, No. 2. pp. 2476-2485.
@article{aae7dd20f61448b8a017833ede8bf991,
title = "Accurate WiFi localization by unsupervised fusion of extended candidate location set",
abstract = "Fusing the predictions of multiple received signal strength (RSS)-based classifiers is an efficient strategy to mitigate the impact of the fluctuation of the RSS. However, most of the existing fusion methods exhibit two remarkable shortcomings: 1) they need to train and store offline weights by the supervised learning and 2) they directly fuse the so-called candidate location set (CLS), which is collected from the most likely location estimate of each classifier (location with the largest probability of being the true location predicted by the classifier), and thus do not fully leverage the knowledge of classifiers. In general, the fluctuation of RSS does not guarantee the location predicted by each classifier with the highest probability to be the true location, thus leading to severe performance degeneration of the existing fusion methods. To overcome the above shortcomings, we propose an accurate WiFi localization framework by unsupervised fusion of an extended CLS (ECLS). First, we train multiple classifiers by only using RSS fingerprints in the offline phase. In the online phase, instead of collecting the CLS from the trained classifiers, we construct an ECLS by augmenting CLS with other location estimates (locations with predication probability greater than a certain threshold) from each classifier. As compared with the CLS, ECLS provides a bigger fusion space that likely includes the true location of the user. Furthermore, an unsupervised fusion localization algorithm based on the ECLS is derived from the joint optimization of weights and the location of the user. Furthermore, a point of inflection searching algorithm is also proposed to intelligently construct the ECLS. Real experimental results show that our proposed algorithm is more robust to changing environments and model errors, and can significantly improve the localization accuracy without any fingerprint and hardware calibrations.",
keywords = "Candidate location set (CLS), WiFi, extended candidate location set (ECLS), indoor localization, received signal strength (RSS), unsupervised fusion",
author = "Xiansheng Guo and Shilin Zhu and Lin Li and Fangzi Hu and Nirwan Ansari",
year = "2019",
month = "4",
day = "1",
doi = "https://doi.org/10.1109/JIOT.2018.2870659",
language = "English (US)",
volume = "6",
pages = "2476--2485",
journal = "IEEE Internet of Things Journal",
issn = "2327-4662",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "2",

}

Accurate WiFi localization by unsupervised fusion of extended candidate location set. / Guo, Xiansheng; Zhu, Shilin; Li, Lin; Hu, Fangzi; Ansari, Nirwan.

In: IEEE Internet of Things Journal, Vol. 6, No. 2, 8466583, 01.04.2019, p. 2476-2485.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Accurate WiFi localization by unsupervised fusion of extended candidate location set

AU - Guo, Xiansheng

AU - Zhu, Shilin

AU - Li, Lin

AU - Hu, Fangzi

AU - Ansari, Nirwan

PY - 2019/4/1

Y1 - 2019/4/1

N2 - Fusing the predictions of multiple received signal strength (RSS)-based classifiers is an efficient strategy to mitigate the impact of the fluctuation of the RSS. However, most of the existing fusion methods exhibit two remarkable shortcomings: 1) they need to train and store offline weights by the supervised learning and 2) they directly fuse the so-called candidate location set (CLS), which is collected from the most likely location estimate of each classifier (location with the largest probability of being the true location predicted by the classifier), and thus do not fully leverage the knowledge of classifiers. In general, the fluctuation of RSS does not guarantee the location predicted by each classifier with the highest probability to be the true location, thus leading to severe performance degeneration of the existing fusion methods. To overcome the above shortcomings, we propose an accurate WiFi localization framework by unsupervised fusion of an extended CLS (ECLS). First, we train multiple classifiers by only using RSS fingerprints in the offline phase. In the online phase, instead of collecting the CLS from the trained classifiers, we construct an ECLS by augmenting CLS with other location estimates (locations with predication probability greater than a certain threshold) from each classifier. As compared with the CLS, ECLS provides a bigger fusion space that likely includes the true location of the user. Furthermore, an unsupervised fusion localization algorithm based on the ECLS is derived from the joint optimization of weights and the location of the user. Furthermore, a point of inflection searching algorithm is also proposed to intelligently construct the ECLS. Real experimental results show that our proposed algorithm is more robust to changing environments and model errors, and can significantly improve the localization accuracy without any fingerprint and hardware calibrations.

AB - Fusing the predictions of multiple received signal strength (RSS)-based classifiers is an efficient strategy to mitigate the impact of the fluctuation of the RSS. However, most of the existing fusion methods exhibit two remarkable shortcomings: 1) they need to train and store offline weights by the supervised learning and 2) they directly fuse the so-called candidate location set (CLS), which is collected from the most likely location estimate of each classifier (location with the largest probability of being the true location predicted by the classifier), and thus do not fully leverage the knowledge of classifiers. In general, the fluctuation of RSS does not guarantee the location predicted by each classifier with the highest probability to be the true location, thus leading to severe performance degeneration of the existing fusion methods. To overcome the above shortcomings, we propose an accurate WiFi localization framework by unsupervised fusion of an extended CLS (ECLS). First, we train multiple classifiers by only using RSS fingerprints in the offline phase. In the online phase, instead of collecting the CLS from the trained classifiers, we construct an ECLS by augmenting CLS with other location estimates (locations with predication probability greater than a certain threshold) from each classifier. As compared with the CLS, ECLS provides a bigger fusion space that likely includes the true location of the user. Furthermore, an unsupervised fusion localization algorithm based on the ECLS is derived from the joint optimization of weights and the location of the user. Furthermore, a point of inflection searching algorithm is also proposed to intelligently construct the ECLS. Real experimental results show that our proposed algorithm is more robust to changing environments and model errors, and can significantly improve the localization accuracy without any fingerprint and hardware calibrations.

KW - Candidate location set (CLS)

KW - WiFi

KW - extended candidate location set (ECLS)

KW - indoor localization

KW - received signal strength (RSS)

KW - unsupervised fusion

UR - http://www.scopus.com/inward/record.url?scp=85053292843&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85053292843&partnerID=8YFLogxK

U2 - https://doi.org/10.1109/JIOT.2018.2870659

DO - https://doi.org/10.1109/JIOT.2018.2870659

M3 - Article

VL - 6

SP - 2476

EP - 2485

JO - IEEE Internet of Things Journal

JF - IEEE Internet of Things Journal

SN - 2327-4662

IS - 2

M1 - 8466583

ER -