Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • 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
Performance improvement of NN based RTLS by customization of NN structure: Heuristic approach
Blekinge Institute of Technology, Faculty of Engineering, Department of Applied Signal Processing.
2015 (English)In: Proceedings of the International Conference on Sensing Technology, ICST, IEEE Computer Society, 2015, p. 278-283Conference paper, Published paper (Refereed)
Abstract [en]

The purpose of this research is to improve performance of the Hybrid Scene Analysis - Neural Network indoor localization algorithm applied in Real-time Locating System, RTLS. A properly customized structure of Neural Network and training algorithms for specific operating environment will enhance the system’s performance in terms of localization accuracy and precision. Due to nonlinearity and model complexity, a heuristic analysis is suitable to evaluate NN performance for different environmental conditions. Efficiency of the proposed customization of a Neural Network is verified by simulations and validated by physical experiments. This research also concerns the influence of size of Neural Network training set. The results prove that, better localization accuracy is with a NN system which is properly customized with respect to a training method, number of neurons and type of transfer function in the hidden layer and also type of transfer function in the output layer.

Place, publisher, year, edition, pages
IEEE Computer Society, 2015. p. 278-283
Series
Proceedings of the International Conference on Sensing Technology, ICST, ISSN 2156-8065
Keywords [en]
Algorithms; Complex networks; Heuristic methods; Network architecture; Neural networks; Optimization; Radio frequency identification (RFID); RSS; Transfer functions, Accuracy and precision; Environmental conditions; Indoor localization; Localization accuracy; Neural network training; Operating environment; Real-Time Locating Systems; Scene analysis, Indoor positioning systems
National Category
Signal Processing
Identifiers
URN: urn:nbn:se:bth-13146DOI: 10.1109/ICSensT.2015.7438407ISI: 000380410400054Scopus ID: 2-s2.0-84964882100ISBN: 9781479963140 (electronic)OAI: oai:DiVA.org:bth-13146DiVA, id: diva2:1016257
Conference
9th International Conference on Sensing Technology, ICST 2015; Auckland; New Zealand
Note

Conference of 9th International Conference on Sensing Technology, ICST 2015 ; Conference Date: 8 December 2015 Through 11 December 2015; Conference Code:121054

Available from: 2016-10-04 Created: 2016-10-03 Last updated: 2019-03-28Bibliographically approved
In thesis
1. Real-time Locating Systems for indoor applications: the methodological customization approach
Open this publication in new window or tab >>Real-time Locating Systems for indoor applications: the methodological customization approach
2019 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

Emerging wireless technologies increase the potential and effectiveness of wireless Real-time Locating Systems (RTLSs), which precisely localize the position, and identify things and people in real time. Among many applications, RTLSs are widely used in the industrial sector for indoor logistics and safety applications. However, signal interferences, which affect the system’s performance, are a serious issue of all indoor RTLS applications. Among others, the interferences are caused by the changeable working environment, the geometry and structure of the space, furnishing, and other obstacles. A customization of the RTLS’s architecture and localization algorithm may provide a way to overcome the interference problem and then enhance the systems’ performance.

The objective of this thesis is to develop and implement customization methods, which enhance system performance in the changeable working environment without compromising the functional and non-functional requirements defined by future users and stakeholders. The customized solution is to be based on the comprehensive methodological analysis of the system’s technical and environmental constraints, along with the requirements specified by the application field. The customization process covers the selection, adjustment and adaptation of the wireless technologies and methods in order to enhance the location system’s performance, in terms of accuracy and precision without compromising its simplicity and price.

In this research, wireless technologies of Radio Frequency Identification (RFID) and Ultra-wideband (UWB) are applied. The related indoor localization methods, such as, ranging techniques based on Received Signal Strength (RSS) and Angle of Arrival (AoA), are a thesis focus. Moreover, estimation methods like Fingerprinting and Angulation are used.

One of the proposed customization methods of RFID-based 3D RTLS, refers to the heuristic analysis-based optimization of a number and configuration of readers. For the same type of system, an alternative way of performance improvement is a customization of localization algorithm, explicitly the Neural Network-based estimation algorithm and its structural features and training methods.

Also in this thesis, performance improvement methods of the AoA-based RTLS operating in an UWB technology are proposed. The proposed customization of this system type is based on the uncertainty pattern defined by a statistical uncertainty model, which maps the localization uncertainty in terms of precision in the 2D workspace. The model depicts how the localization uncertainty depends on an arrangement of Location Sensors and workspace geometry. Another proposed customization method is realized by defining and implementing correction vectors for different working environments, which enhance the system’s performance in terms of its accuracy.

This thesis consists of two parts. Part I, Prolegomena, presents the overview of applied theories and research methods. This part aims to illustrate the links between the articles constituting the second part of the dissertation. Part II, Papers consists of five reformatted papers already published in peer reviewed journals and conferences.

Place, publisher, year, edition, pages
Karlskrona: Blekinge Tekniska Högskola, 2019. p. 204
Series
Blekinge Institute of Technology Doctoral Dissertation Series, ISSN 1653-2090 ; 08
Keywords
Accuracy and Precision, Angle of Arrival, Fingerprinting Method, Indoor Localization, Indoor Positioning System, Multi-Sensor System, Neural Network, Radio Frequency Identification - RFID, Real Time Locating System, Received Signal Strength, RFID Network Planning, Scene Analysis, Sensors Arrangement, System Customization, Uncertainty, Uncertainty Map, User Driven Design
National Category
Signal Processing
Identifiers
urn:nbn:se:bth-17740 (URN)978-91-7295-373-4 (ISBN)
Public defence
2019-05-03, J1650, Blekinge Tekniska Högskola, Campus Gräsvik, Karlskrona, 13:15 (English)
Opponent
Supervisors
Available from: 2019-03-28 Created: 2019-03-25 Last updated: 2019-06-18Bibliographically approved

Open Access in DiVA

No full text in DiVA

Other links

Publisher's full textScopus

Authority records BETA

Kulesza, Wlodek

Search in DiVA

By author/editor
Kulesza, Wlodek
By organisation
Department of Applied Signal Processing
Signal Processing

Search outside of DiVA

GoogleGoogle Scholar

doi
isbn
urn-nbn

Altmetric score

doi
isbn
urn-nbn
Total: 118 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • 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