Random forest learning method to identify different objects using channel estimations from VLC link

Mehmet C. Ilter; Alexis A. Dowhuszko; Kiran K. Vangapattu; Kubra Kutlu; Jyri Hämäläinen

doi:10.1109/PIMRC48278.2020.9217131

Random forest learning method to identify different objects using channel estimations from VLC link

Mehmet C. Ilter, Alexis A. Dowhuszko, Kiran K. Vangapattu, Kubra Kutlu, Jyri Hämäläinen

Centre Tecnològic de Telecomunicacions de Catalunya

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

4 Citations (Scopus)

173 Downloads (Pure)

Abstract

This paper demonstrates the feasibility of using supervised learning algorithms to identify the presence of different objects, taking advantage of the effect that they create on the VLC channel gains. For this purpose, a software-defined VLC link is implemented using a Phosphor-converted LED, whose light intensity is modulated by an Optical OFDM frame that includes synchronization words and pilot sequences for channel estimation. Actual estimated channel gains are collected in the receiver, which are used to train and assess the performance of the Random Forest classifier. The accuracy of the monitoring system is evaluated using three different objects, showing an accuracy in the order of 90% in detecting the objects, even when they take different positions when obstructing the VLC link.

Original language	English
Title of host publication	Proceedings of the IEEE 31th Annual International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC 2020
Publisher	IEEE
Number of pages	6
ISBN (Electronic)	9781728144900
DOIs	https://doi.org/10.1109/PIMRC48278.2020.9217131
Publication status	Published - Aug 2020
MoE publication type	A4 Conference publication
Event	IEEE International Symposium on Personal, Indoor and Mobile Radio Communications - Virtual, Online Duration: 31 Aug 2020 → 3 Sept 2020

Publication series

Name	IEEE International Symposium on Personal, Indoor, and Mobile Radio Communications workshops
Publisher	IEEE
ISSN (Print)	2166-9570
ISSN (Electronic)	2166-9589

Conference

Conference	IEEE International Symposium on Personal, Indoor and Mobile Radio Communications
Abbreviated title	PIMRC
City	Virtual, Online
Period	31/08/2020 → 03/09/2020
Other	Virtual conference

Keywords

Indoor monitoring
Optical OFDM
Random Forest
Software-defined
Supervised Learning
VLC system

Access to Document

10.1109/PIMRC48278.2020.9217131

ELEC_Ilter_etal_Random-Forest-Learning_PIMRC2020Accepted author manuscript, 1.06 MB

Cite this

Ilter, M. C., Dowhuszko, A. A., Vangapattu, K. K., Kutlu, K., & Hämäläinen, J. (2020). Random forest learning method to identify different objects using channel estimations from VLC link. In Proceedings of the IEEE 31th Annual International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC 2020 Article 9217131 (IEEE International Symposium on Personal, Indoor, and Mobile Radio Communications workshops). IEEE. https://doi.org/10.1109/PIMRC48278.2020.9217131

Ilter, Mehmet C. ; Dowhuszko, Alexis A. ; Vangapattu, Kiran K. et al. / Random forest learning method to identify different objects using channel estimations from VLC link. Proceedings of the IEEE 31th Annual International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC 2020. IEEE, 2020. (IEEE International Symposium on Personal, Indoor, and Mobile Radio Communications workshops).

@inproceedings{99fda2a23112406db696cd3c83d37de1,

title = "Random forest learning method to identify different objects using channel estimations from VLC link",

abstract = "This paper demonstrates the feasibility of using supervised learning algorithms to identify the presence of different objects, taking advantage of the effect that they create on the VLC channel gains. For this purpose, a software-defined VLC link is implemented using a Phosphor-converted LED, whose light intensity is modulated by an Optical OFDM frame that includes synchronization words and pilot sequences for channel estimation. Actual estimated channel gains are collected in the receiver, which are used to train and assess the performance of the Random Forest classifier. The accuracy of the monitoring system is evaluated using three different objects, showing an accuracy in the order of 90% in detecting the objects, even when they take different positions when obstructing the VLC link.",

keywords = "Indoor monitoring, Optical OFDM, Random Forest, Software-defined, Supervised Learning, VLC system",

author = "Ilter, {Mehmet C.} and Dowhuszko, {Alexis A.} and Vangapattu, {Kiran K.} and Kubra Kutlu and Jyri H{\"a}m{\"a}l{\"a}inen",

note = "| openaire: EC/H2020/777222/EU//ATTRACT; IEEE International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC ; Conference date: 31-08-2020 Through 03-09-2020",

year = "2020",

month = aug,

doi = "10.1109/PIMRC48278.2020.9217131",

language = "English",

series = "IEEE International Symposium on Personal, Indoor, and Mobile Radio Communications workshops",

publisher = "IEEE",

booktitle = "Proceedings of the IEEE 31th Annual International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC 2020",

address = "United States",

}

Ilter, MC , Dowhuszko, AA, Vangapattu, KK, Kutlu, K & Hämäläinen, J 2020, Random forest learning method to identify different objects using channel estimations from VLC link. in Proceedings of the IEEE 31th Annual International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC 2020., 9217131, IEEE International Symposium on Personal, Indoor, and Mobile Radio Communications workshops, IEEE, IEEE International Symposium on Personal, Indoor and Mobile Radio Communications, Virtual, Online, 31/08/2020. https://doi.org/10.1109/PIMRC48278.2020.9217131

Random forest learning method to identify different objects using channel estimations from VLC link. / Ilter, Mehmet C.; Dowhuszko, Alexis A.; Vangapattu, Kiran K. et al.
Proceedings of the IEEE 31th Annual International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC 2020. IEEE, 2020. 9217131 (IEEE International Symposium on Personal, Indoor, and Mobile Radio Communications workshops).

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

TY - GEN

T1 - Random forest learning method to identify different objects using channel estimations from VLC link

AU - Ilter, Mehmet C.

AU - Dowhuszko, Alexis A.

AU - Vangapattu, Kiran K.

AU - Kutlu, Kubra

AU - Hämäläinen, Jyri

N1 - | openaire: EC/H2020/777222/EU//ATTRACT

PY - 2020/8

Y1 - 2020/8

N2 - This paper demonstrates the feasibility of using supervised learning algorithms to identify the presence of different objects, taking advantage of the effect that they create on the VLC channel gains. For this purpose, a software-defined VLC link is implemented using a Phosphor-converted LED, whose light intensity is modulated by an Optical OFDM frame that includes synchronization words and pilot sequences for channel estimation. Actual estimated channel gains are collected in the receiver, which are used to train and assess the performance of the Random Forest classifier. The accuracy of the monitoring system is evaluated using three different objects, showing an accuracy in the order of 90% in detecting the objects, even when they take different positions when obstructing the VLC link.

AB - This paper demonstrates the feasibility of using supervised learning algorithms to identify the presence of different objects, taking advantage of the effect that they create on the VLC channel gains. For this purpose, a software-defined VLC link is implemented using a Phosphor-converted LED, whose light intensity is modulated by an Optical OFDM frame that includes synchronization words and pilot sequences for channel estimation. Actual estimated channel gains are collected in the receiver, which are used to train and assess the performance of the Random Forest classifier. The accuracy of the monitoring system is evaluated using three different objects, showing an accuracy in the order of 90% in detecting the objects, even when they take different positions when obstructing the VLC link.

KW - Indoor monitoring

KW - Optical OFDM

KW - Random Forest

KW - Software-defined

KW - Supervised Learning

KW - VLC system

UR - http://www.scopus.com/inward/record.url?scp=85094141446&partnerID=8YFLogxK

U2 - 10.1109/PIMRC48278.2020.9217131

DO - 10.1109/PIMRC48278.2020.9217131

M3 - Conference article in proceedings

AN - SCOPUS:85094141446

T3 - IEEE International Symposium on Personal, Indoor, and Mobile Radio Communications workshops

BT - Proceedings of the IEEE 31th Annual International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC 2020

PB - IEEE

T2 - IEEE International Symposium on Personal, Indoor and Mobile Radio Communications

Y2 - 31 August 2020 through 3 September 2020

ER -

Ilter MC , Dowhuszko AA, Vangapattu KK, Kutlu K, Hämäläinen J. Random forest learning method to identify different objects using channel estimations from VLC link. In Proceedings of the IEEE 31th Annual International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC 2020. IEEE. 2020. 9217131. (IEEE International Symposium on Personal, Indoor, and Mobile Radio Communications workshops). doi: 10.1109/PIMRC48278.2020.9217131

Random forest learning method to identify different objects using channel estimations from VLC link

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

ATTRACT: breAkThrough innovaTion pRogrAmme for a pan-European Detection and Imaging eCosysTem

Cite this

Random forest learning method to identify different objects using channel estimations from VLC link

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Projects

ATTRACT: breAkThrough innovaTion pRogrAmme for a pan-European Detection and Imaging eCosysTem

Cite this