liu.seSearch for publications in DiVA
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • oxford
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Gaussian Processes for Positioning Using Radio Signal Strength Measurements
Linköping University, Department of Electrical Engineering, Automatic Control. Linköping University, Faculty of Science & Engineering.ORCID iD: 0000-0003-1214-2391
2019 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

Estimation of unknown parameters is considered as one of the major research areas in statistical signal processing. In the most recent decades, approaches in estimation theory have become more and more attractive in practical applications. Examples of such applications may include, but are not limited to, positioning using various measurable radio signals in indoor environments, self-navigation for autonomous cars, image processing, radar tracking and so on. One issue that is usually encountered when solving an estimation problem is to identify a good system model, which may have great impacts on the estimation performance. In this thesis, we are interested in studying estimation problems particularly in inferring the unknown positions from noisy radio signal measurements. In addition, the modeling of the system is studied by investigating the relationship between positions and radio signal strength measurements.

One of the main contributions of this thesis is to propose a novel indoor positioning framework based on proximity measurements, which are obtained by quantizing the received signal strength measurements. Sequential Monte Carlo methods, to be more specific particle filter and smoother, are utilized for estimating unknown positions from proximity measurements. The Cramér-Rao bounds for proximity-based positioning are further derived as a benchmark for the positioning accuracy in this framework.

Secondly, to improve the estimation performance, Bayesian non-parametric modeling, namely Gaussian processes, have been adopted to provide more accurate and flexible models for both dynamic motions and radio signal strength measurements. Then, the Cramér-Rao bounds for Gaussian process based system models are derived and evaluated in an indoor positioning scenario.

In addition, we estimate the positions of stationary devices by comparing the individual signal strength measurements with a pre-constructed fingerprinting database. The positioning accuracy is further compared to the case where a moving device is positioned using a time series of radio signal strength measurements.

Moreover, Gaussian processes have been applied to sports analytics, where trajectory modeling for athletes is studied. The proposed framework can be further utilized to carry out, for instance, performance prediction and analysis, health condition monitoring, etc. Finally, a grey-box modeling is proposed to analyze the forces, particularly in cross-country skiing races, by combining a deterministic kinetic model with Gaussian process.

Place, publisher, year, edition, pages
Linköping: Linköping University Electronic Press, 2019. , p. 51
Series
Linköping Studies in Science and Technology. Dissertations, ISSN 0345-7524 ; 1968
Keywords [en]
Gaussian process, positioning, radio signals
National Category
Signal Processing
Identifiers
URN: urn:nbn:se:liu:diva-153944DOI: 10.3384/diss.diva-153944ISBN: 978-91-7685-162-3 (print)OAI: oai:DiVA.org:liu-153944DiVA, id: diva2:1288029
Public defence
2019-03-15, Ada Lovelace, Campus Valla, Linköping, 10:15 (English)
Opponent
Supervisors
Funder
ELLIIT - The Linköping‐Lund Initiative on IT and Mobile CommunicationsAvailable from: 2019-02-27 Created: 2019-02-12 Last updated: 2019-02-27Bibliographically approved
List of papers
1. Received-Signal-Strength Threshold Optimization Using Gaussian Processes
Open this publication in new window or tab >>Received-Signal-Strength Threshold Optimization Using Gaussian Processes
2017 (English)In: IEEE Transactions on Signal Processing, ISSN 1053-587X, E-ISSN 1941-0476, Vol. 65, no 8, p. 2164-2177Article in journal (Refereed) Published
Abstract [en]

There is a big trend nowadays to use event-triggered proximity report for indoor positioning. This paper presents a generic received-signal-strength (RSS) threshold optimization framework for generating informative proximity reports. The proposed framework contains five main building blocks, namely the deployment information, RSS model, positioning metric selection, optimization process and management. Among others, we focus on Gaussian process regression (GPR)-based RSS models and positioning metric computation. The optimal RSS threshold is found through minimizing the best achievable localization root-mean-square-error formulated with the aid of fundamental lower bound analysis. Computational complexity is compared for different RSS models and different fundamental lower bounds. The resulting optimal RSS threshold enables enhanced performance of new fashioned low-cost and low-complex proximity report-based positioning algorithms. The proposed framework is validated with real measurements collected in an office area where bluetooth-low-energy (BLE) beacons are deployed.

Place, publisher, year, edition, pages
Institute of Electrical and Electronics Engineers (IEEE), 2017
National Category
Electrical Engineering, Electronic Engineering, Information Engineering
Identifiers
urn:nbn:se:liu:diva-135065 (URN)10.1109/TSP.2017.2655480 (DOI)000395827100018 ()
Projects
TRAX
Note

Funding agencies: European Union FP7 Marie Curie training programme on Tracking in Complex Sensor Systems [607400]

Available from: 2017-03-08 Created: 2017-03-08 Last updated: 2019-02-12Bibliographically approved
2. Sequential Monte Carlo Methods and Theoretical Bounds for Proximity Report Based Indoor Positioning
Open this publication in new window or tab >>Sequential Monte Carlo Methods and Theoretical Bounds for Proximity Report Based Indoor Positioning
Show others...
2018 (English)In: IEEE Transactions on Vehicular Technology, ISSN 0018-9545, E-ISSN 1939-9359, Vol. 67, no 6, p. 5372-5386Article in journal (Refereed) Published
Abstract [en]

The commercial interest in proximity services is increasing. Application examples include location-based information and advertisements, logistics, social networking, file sharing, etc. In this paper, we consider positioning of devices based on a time series of proximity reports from a mobile device to a network node. This corresponds to nonlinear measurements with respect to the device position in relation to the network nodes. Motion model will be needed together with the measurements to determine the position of the device. Therefore, sequential Monte Carlo methods, namely particle filtering and smoothing, are applicable for positioning. Positioning performance is evaluated in a typical office area with Bluetooth-low-energy beacons deployed for proximity detection and report, and is further compared to parametric Cramér-Rao lower bounds. Finally, the position accuracy is also evaluated with real experimental data.

Place, publisher, year, edition, pages
Institute of Electrical and Electronics Engineers (IEEE), 2018
Keywords
Proximity, indoor positioning, particle filtering and smoothing, Cramer-Rao lower bounds
National Category
Signal Processing
Identifiers
urn:nbn:se:liu:diva-147834 (URN)10.1109/TVT.2018.2799174 (DOI)000435553400053 ()2-s2.0-85041415767 (Scopus ID)
Note

Funding agencies: European Union FP7 Marie Curie Training Programme on Tracking in Complex Sensor Systems (TRAX) [607400]; NSFC [61701426]; Shenzhen Science and Technology Innovation Council [JCYJ20170307155957688, JCYJ20170411102101881]

Available from: 2018-05-15 Created: 2018-05-15 Last updated: 2019-02-12Bibliographically approved
3. Gaussian Process for Propagation modeling and Proximity Reports Based Indoor Positioning
Open this publication in new window or tab >>Gaussian Process for Propagation modeling and Proximity Reports Based Indoor Positioning
Show others...
2016 (English)In: 2016 IEEE 83rd Vehicular Technology Conference (VTC Spring), IEEE , 2016, p. 1-5Conference paper, Published paper (Refereed)
Abstract [en]

The commercial interest in proximity services is increasing. Application examples include location-based information and advertisements, logistics, social networking, file sharing, etc. In this paper, we consider network-based positioning based on times series of proximity reports from a mobile device, either only a proximity indicator, or a vector of RSS from observed nodes. Such positioning corresponds to a latent and nonlinear observation model. To address these problems, we combine two powerful tools, namely particle filtering and Gaussian process regression (GPR) for radio signal propagation modeling. The latter also provides some insights into the spatial correlation of the radio propagation in the considered area. Radio propagation modeling and positioning performance are evaluated in a typical office area with Bluetooth-Low-Energy (BLE) beacons deployed for proximity detection and reports. Results show that the positioning accuracy can be improved by using GPR.

Place, publisher, year, edition, pages
IEEE, 2016
National Category
Communication Systems
Identifiers
urn:nbn:se:liu:diva-128255 (URN)10.1109/VTCSpring.2016.7504255 (DOI)000386528400206 ()9781509016983 (ISBN)
Conference
2016 IEEE 83rd Vehicular Technology Conference: VTC2016-Spring, 15–18 May 2016, Nanjing, China
Available from: 2016-05-24 Created: 2016-05-24 Last updated: 2019-02-12Bibliographically approved
4. Gaussian processes for RSS fingerprints construction in indoor localization
Open this publication in new window or tab >>Gaussian processes for RSS fingerprints construction in indoor localization
2018 (English)In: 21st International Conference on Information Fusion (FUSION), IEEE, 2018, p. 1377-1384Conference paper, Published paper (Refereed)
Abstract [en]

Location-based applications attract more and more attention in recent years. Examples of such applications include commercial advertisements, social networking software and patient monitoring. The received signal strength (RSS) based location fingerprinting is one of the most popular solutions for indoor localization. However, there is a big challenge in collecting and maintaining a relatively large RSS fingerprint database. In this work, we propose and compare two algorithms namely, the Gaussian process (GP) and Gaussian process with variogram, to estimate and construct the RSS fingerprints with incomplete data. The fingerprint of unknown reference points is estimated based on measurements at a limited number of surrounding locations. To validate the effectiveness of both algorithms, experiments using Bluetooth-low-energy (BLE) infrastructure have been conducted. The constructed RSS fingerprints are compared to the true measurements, and the result is analyzed. Finally, using the constructed fingerprints, the localization performance of a probabilistic fingerprinting method is evaluated.

Place, publisher, year, edition, pages
IEEE, 2018
National Category
Signal Processing
Identifiers
urn:nbn:se:liu:diva-151693 (URN)10.23919/ICIF.2018.8455842 (DOI)978-0-9964527-6-2 (ISBN)
Conference
21st International Conference on Information Fusion (FUSION), 10-13 July 2018, Cambridge, UK
Available from: 2018-10-01 Created: 2018-10-01 Last updated: 2019-02-12
5. Gaussian Processes for Flow Modeling and Prediction of Positioned Trajectories Evaluated with Sports Data
Open this publication in new window or tab >>Gaussian Processes for Flow Modeling and Prediction of Positioned Trajectories Evaluated with Sports Data
Show others...
2016 (English)In: 19th International Conference on  Information Fusion (FUSION), 2016, Institute of Electrical and Electronics Engineers (IEEE), 2016, p. 1461-1468Conference paper, Published paper (Refereed)
Abstract [en]

Kernel-based machine learning methods are gaining increasing interest in flow modeling and prediction in recent years. Gaussian process (GP) is one example of such kernelbased methods, which can provide very good performance for nonlinear problems. In this work, we apply GP regression to flow modeling and prediction of athletes in ski races, but the proposed framework can be generally applied to other use cases with device trajectories of positioned data. Some specific aspects can be addressed when the data is periodic, like in sports where the event is split up over multiple laps along a specific track. Flow models of both the individual skier and a cluster of skiers are derived and analyzed. Performance has been evaluated using data from the Falun Nordic World Ski Championships 2015, in particular the Men’s cross country 4 × 10 km relay. The results show that the flow models vary spatially for different skiers and clusters. We further demonstrate that GP regression provides powerful and accurate models for flow prediction.

Place, publisher, year, edition, pages
Institute of Electrical and Electronics Engineers (IEEE), 2016
National Category
Electrical Engineering, Electronic Engineering, Information Engineering Probability Theory and Statistics
Identifiers
urn:nbn:se:liu:diva-129758 (URN)9780996452748 (ISBN)9781509020126 (ISBN)
Conference
19th International Conference on Information Fusion, 5-8 July 2016, Heidelberg, Germany
Available from: 2016-06-27 Created: 2016-06-27 Last updated: 2019-02-12Bibliographically approved

Open Access in DiVA

Gaussian Processes for Positioning Using Radio Signal Strength Measurements(626 kB)176 downloads
File information
File name FULLTEXT02.pdfFile size 626 kBChecksum SHA-512
fed9804cc4f62e753d01a7b6e4634fb02a15d5e8ba7bebb48659226d76f2583824e55a497cd49c226d05093d66608a1d6bde49f51d14b1b2b92060a56e46ce57
Type fulltextMimetype application/pdf

Other links

Publisher's full text

Authority records BETA

Zhao, Yuxin

Search in DiVA

By author/editor
Zhao, Yuxin
By organisation
Automatic ControlFaculty of Science & Engineering
Signal Processing

Search outside of DiVA

GoogleGoogle Scholar
Total: 176 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

doi
isbn
urn-nbn

Altmetric score

doi
isbn
urn-nbn
Total: 1221 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • oxford
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf