Mid Sweden University

miun.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • 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
Machine Learning-Aided Classification of LoS/NLoS Radio Links in Industrial IoT
Mid Sweden University, Faculty of Science, Technology and Media, Department of Information Systems and Technology.
Mid Sweden University, Faculty of Science, Technology and Media, Department of Information Systems and Technology.
Mid Sweden University, Faculty of Science, Technology and Media, Department of Information Systems and Technology.ORCID iD: 0000-0003-3717-7793
Mid Sweden University, Faculty of Science, Technology and Media, Department of Information Systems and Technology.ORCID iD: 0000-0003-0873-7827
2020 (English)In: 2020 16th IEEE International Conference on Factory Communication Systems (WFCS), IEEE, 2020Conference paper, Published paper (Refereed)
Abstract [en]

Wireless sensors and actuators networks are an essential element to realize industrial IoT (IIoT) systems, yet their diffusion is hampered by the complexity of ensuring reliable communication in industrial environments.A significant problem with that respect is the unpredictable fluctuation of a radio-link between the line-of-sight (LoS) and the non-line-of-sight (NLoS) state due to time-varying environments.The impact of link-state over reception performance, suggests that link-state variations should be monitored at run-time, enabling dynamic adaptation of the transmission scheme on a link-basis to safeguard QoS.Starting from the assumption that accurate channel-sounding is unsuitable for low-complexity IIoT devices, we investigate the feasibility of channel-state identification for platforms with limited sensing capabilities. In this context, we evaluate the performance of different supervised-learning algorithms with variable complexity for the inference of the radio-link state.Our approach provides fast link-diagnostics by performing online classification based on a single received packet. Furthermore, the method takes into account the effects of limited sampling frequency, bit-depth, and moving average filtering, which are typical to hardware-constrained platforms.The results of an experimental campaign in both industrial and office environments show promising classification accuracy of LoS/NLoS radio links. Additional tests indicate that the proposed method retains good performance even with low-resolution RSSI-samples available in low-cost WSN nodes, which facilitates its adoption in real IIoT networks.

Place, publisher, year, edition, pages
IEEE, 2020.
National Category
Communication Systems
Identifiers
URN: urn:nbn:se:miun:diva-39031DOI: 10.1109/WFCS47810.2020.9114409ISI: 000847108800016Scopus ID: 2-s2.0-85089105584ISBN: 978-1-7281-5297-4 (electronic)OAI: oai:DiVA.org:miun-39031DiVA, id: diva2:1430635
Conference
16th IEEE International Conference on Factory Communication Systems (WFCS 2020), Porto, Portugal / Online, 27-29 April, 2020
Available from: 2020-05-15 Created: 2020-05-15 Last updated: 2022-12-21Bibliographically approved
In thesis
1. Towards Radio-Environment Aware IoT Networks: Wireless Coexistence Methods for Low-complexity Devices
Open this publication in new window or tab >>Towards Radio-Environment Aware IoT Networks: Wireless Coexistence Methods for Low-complexity Devices
2020 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

Wireless technologies for short-range communication play a central role in the massive diffusion of the Internet of Things (IoT) paradigm. Such communication solutions rely extensively on the availability of unlicensed spectrum in the form of bands for industrial, scientific, and medical (ISM) applications. While ISM bands greatly simplify network deployments by avoiding operator-related costs and facilitating worldwide applicability, they present the shortcoming of non-cooperative spectrum usage, which manifests in the form of radio interference. Interference and time-varying environments generate complex and dynamic scenarios for wireless network deployments, endangering communication performance. The problem becomes especially critical when the timeliness and reliability of the communication are subject to stringent requirements, which is the case for several industrial IoT (IIoT) applications.

This work aims to enhance the reliability and performance of wireless communication in IoT networks by enriching the existing methods for radio-environment analysis. The central idea of this research is that a run-time analysis of the radio channel properties is a crucial element to ensure performance stability in unpredictable radio environments with potentially disruptive interference.An added challenge of this work comes from the hypothesis that such an analysis can be performed even with strongly resource-constrained platforms without hindering routine network functionalities. The employed methodology is heavily reliant on experimental validation, encompassing implementation on IoT radio devices and measurement campaigns. This thesis makes two principal scientific contributions.

The first contribution is the design of a comprehensive collection of methods for the analysis of the radio environment, designed to operate entirely onboard on IoT radio platforms.The approaches encompass interference detection, classification, spectrum analysis, link-state analysis, and detection of outages in end-to-end communication. The methods are designed to overcome the gap that exists in the related literature between the elaborate signal analysis operated with dedicated hardware and the lightweight, but sub-optimal, analysis methods developed for legacy wireless sensor networks.

The second contribution of this work is made by showing potential uses of the developed analysis methods to: i) safeguard the performance of wireless communication under interference and ii) enhance the coexistence of co-located wireless networks. To this end, firstly, a proactive method for dynamic blacklisting is designed that exploits real-time signal analysis and significantly improves the communication reliability of an IIoT radio link under heavy radio interference. Secondly, a method for autonomous radio environment mapping (REM) in IoT networks is proposed that employs onboard interference identification and tracks the sources of wireless interference in space, time, and frequency. The approach ensures a dynamic level of REM detail and provides a powerful tool for predicting the IoT network performance and adapting the network parameters at run-time.

Place, publisher, year, edition, pages
Sundsvall: Mid Sweden University, 2020. p. 92
Series
Mid Sweden University doctoral thesis, ISSN 1652-893X ; 323
Keywords
IoT, Industrial IoT, Interference, Machine Learning, Wireless Coexistence, Wireless Networks
National Category
Communication Systems
Identifiers
urn:nbn:se:miun:diva-39034 (URN)978-91-88947-53-6 (ISBN)
Public defence
2020-06-16, Zoom, Holmgatan,10, Sundsvall, 09:00 (English)
Opponent
Supervisors
Funder
Knowledge Foundation
Note

Vid tidpunkten för disputationen var följande delarbete opublicerat: delarbete 7 inskickat.

At the time of the doctoral defence the following paper was unpublished: paper 7 submitted.

Available from: 2020-05-18 Created: 2020-05-15 Last updated: 2020-05-19Bibliographically approved

Open Access in DiVA

No full text in DiVA

Other links

Publisher's full textScopus

Authority records

Bombino, AndreaGrimaldi, SimoneMahmood, AamirGidlund, Mikael

Search in DiVA

By author/editor
Bombino, AndreaGrimaldi, SimoneMahmood, AamirGidlund, Mikael
By organisation
Department of Information Systems and Technology
Communication Systems

Search outside of DiVA

GoogleGoogle Scholar

doi
isbn
urn-nbn

Altmetric score

doi
isbn
urn-nbn
Total: 443 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • 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