TDOA-based localization via stochastic gradient descent variants

L.F. Abanto Leon, Arie G.C. Koppelaar, S.M. Heemstra de Groot

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/CongresprocedureConferentiebijdrageAcademicpeer review

65 Downloads (Pure)

Uittreksel

Source localization is of pivotal importance in several areas such as wireless sensor networks and Internet of Things (IoT), where the location information can be used for a variety of purposes, e.g. surveillance, monitoring, tracking, etc. Time Difference of Arrival (TDOA) is one of the well-known localization approaches where the source broadcasts a signal and a number of receivers record the arriving time of the transmitted signal. By means of computing the time difference from various receivers, the source location can be estimated. On the other hand, in the recent few years novel optimization algorithms have appeared in the literature for $(i)$ processing big data and for $(ii)$ training deep neural networks. Most of these techniques are enhanced variants of the classical stochastic gradient descent (SGD) but with additional features that promote faster convergence. In this paper, we compare the performance of the classical SGD with the novel techniques mentioned above. In addition, we propose an optimization procedure called RMSProp+AF, which is based on RMSProp algorithm but with the advantage of incorporating adaptation of the decaying factor. We show through simulations that all of these techniques---which are commonly used in the machine learning domain---can also be successfully applied to signal processing problems and are capable of attaining improved convergence and stability. Finally, it is also shown through simulations that the proposed method can outperform other competing approaches as both its convergence and stability are superior.
Originele taal-2Engels
TitelIEEE 88th Vehicular Technology Conference
SubtitelVTC 2018-Fall
Plaats van productiePiscataway
UitgeverijInstitute of Electrical and Electronics Engineers
Aantal pagina's11
ISBN van elektronische versie9781538663585
DOI's
StatusGepubliceerd - 15 apr 2019
Evenement88th IEEE Vehicular Technology Conference, VTC-Fall 2018 - Chicago, Verenigde Staten van Amerika
Duur: 27 aug 201830 aug 2018
Congresnummer: 88
http://www.ieeevtc.org/vtc2018fall/

Congres

Congres88th IEEE Vehicular Technology Conference, VTC-Fall 2018
Verkorte titelVTC2018-Fall
LandVerenigde Staten van Amerika
StadChicago
Periode27/08/1830/08/18
Internet adres

Vingerafdruk

Stochastic Gradient
Gradient Descent
Stability and Convergence
Receiver
Learning systems
Wireless sensor networks
Source Localization
Internet of Things
Signal processing
Surveillance
Broadcast
Signal Processing
Wireless Sensor Networks
Monitoring
Optimization Algorithm
Machine Learning
Simulation
Processing
Neural Networks
Optimization

Citeer dit

Abanto Leon, L. F., Koppelaar, A. G. C., & Heemstra de Groot, S. M. (2019). TDOA-based localization via stochastic gradient descent variants. In IEEE 88th Vehicular Technology Conference: VTC 2018-Fall [8690742] Piscataway: Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/VTCFall.2018.8690742
Abanto Leon, L.F. ; Koppelaar, Arie G.C. ; Heemstra de Groot, S.M. / TDOA-based localization via stochastic gradient descent variants. IEEE 88th Vehicular Technology Conference: VTC 2018-Fall. Piscataway : Institute of Electrical and Electronics Engineers, 2019.
@inproceedings{01d1ba75e6cc4218b64402409cb3d857,
title = "TDOA-based localization via stochastic gradient descent variants",
abstract = "Source localization is of pivotal importance in several areas such as wireless sensor networks and Internet of Things (IoT), where the location information can be used for a variety of purposes, e.g. surveillance, monitoring, tracking, etc. Time Difference of Arrival (TDOA) is one of the well-known localization approaches where the source broadcasts a signal and a number of receivers record the arriving time of the transmitted signal. By means of computing the time difference from various receivers, the source location can be estimated. On the other hand, in the recent few years novel optimization algorithms have appeared in the literature for $(i)$ processing big data and for $(ii)$ training deep neural networks. Most of these techniques are enhanced variants of the classical stochastic gradient descent (SGD) but with additional features that promote faster convergence. In this paper, we compare the performance of the classical SGD with the novel techniques mentioned above. In addition, we propose an optimization procedure called RMSProp+AF, which is based on RMSProp algorithm but with the advantage of incorporating adaptation of the decaying factor. We show through simulations that all of these techniques---which are commonly used in the machine learning domain---can also be successfully applied to signal processing problems and are capable of attaining improved convergence and stability. Finally, it is also shown through simulations that the proposed method can outperform other competing approaches as both its convergence and stability are superior.",
author = "{Abanto Leon}, L.F. and Koppelaar, {Arie G.C.} and {Heemstra de Groot}, S.M.",
year = "2019",
month = "4",
day = "15",
doi = "10.1109/VTCFall.2018.8690742",
language = "English",
booktitle = "IEEE 88th Vehicular Technology Conference",
publisher = "Institute of Electrical and Electronics Engineers",
address = "United States",

}

Abanto Leon, LF, Koppelaar, AGC & Heemstra de Groot, SM 2019, TDOA-based localization via stochastic gradient descent variants. in IEEE 88th Vehicular Technology Conference: VTC 2018-Fall., 8690742, Institute of Electrical and Electronics Engineers, Piscataway, 88th IEEE Vehicular Technology Conference, VTC-Fall 2018, Chicago, Verenigde Staten van Amerika, 27/08/18. https://doi.org/10.1109/VTCFall.2018.8690742

TDOA-based localization via stochastic gradient descent variants. / Abanto Leon, L.F.; Koppelaar, Arie G.C.; Heemstra de Groot, S.M.

IEEE 88th Vehicular Technology Conference: VTC 2018-Fall. Piscataway : Institute of Electrical and Electronics Engineers, 2019. 8690742.

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/CongresprocedureConferentiebijdrageAcademicpeer review

TY - GEN

T1 - TDOA-based localization via stochastic gradient descent variants

AU - Abanto Leon, L.F.

AU - Koppelaar, Arie G.C.

AU - Heemstra de Groot, S.M.

PY - 2019/4/15

Y1 - 2019/4/15

N2 - Source localization is of pivotal importance in several areas such as wireless sensor networks and Internet of Things (IoT), where the location information can be used for a variety of purposes, e.g. surveillance, monitoring, tracking, etc. Time Difference of Arrival (TDOA) is one of the well-known localization approaches where the source broadcasts a signal and a number of receivers record the arriving time of the transmitted signal. By means of computing the time difference from various receivers, the source location can be estimated. On the other hand, in the recent few years novel optimization algorithms have appeared in the literature for $(i)$ processing big data and for $(ii)$ training deep neural networks. Most of these techniques are enhanced variants of the classical stochastic gradient descent (SGD) but with additional features that promote faster convergence. In this paper, we compare the performance of the classical SGD with the novel techniques mentioned above. In addition, we propose an optimization procedure called RMSProp+AF, which is based on RMSProp algorithm but with the advantage of incorporating adaptation of the decaying factor. We show through simulations that all of these techniques---which are commonly used in the machine learning domain---can also be successfully applied to signal processing problems and are capable of attaining improved convergence and stability. Finally, it is also shown through simulations that the proposed method can outperform other competing approaches as both its convergence and stability are superior.

AB - Source localization is of pivotal importance in several areas such as wireless sensor networks and Internet of Things (IoT), where the location information can be used for a variety of purposes, e.g. surveillance, monitoring, tracking, etc. Time Difference of Arrival (TDOA) is one of the well-known localization approaches where the source broadcasts a signal and a number of receivers record the arriving time of the transmitted signal. By means of computing the time difference from various receivers, the source location can be estimated. On the other hand, in the recent few years novel optimization algorithms have appeared in the literature for $(i)$ processing big data and for $(ii)$ training deep neural networks. Most of these techniques are enhanced variants of the classical stochastic gradient descent (SGD) but with additional features that promote faster convergence. In this paper, we compare the performance of the classical SGD with the novel techniques mentioned above. In addition, we propose an optimization procedure called RMSProp+AF, which is based on RMSProp algorithm but with the advantage of incorporating adaptation of the decaying factor. We show through simulations that all of these techniques---which are commonly used in the machine learning domain---can also be successfully applied to signal processing problems and are capable of attaining improved convergence and stability. Finally, it is also shown through simulations that the proposed method can outperform other competing approaches as both its convergence and stability are superior.

UR - http://www.scopus.com/inward/record.url?scp=85064910908&partnerID=8YFLogxK

U2 - 10.1109/VTCFall.2018.8690742

DO - 10.1109/VTCFall.2018.8690742

M3 - Conference contribution

AN - SCOPUS:85064910908

BT - IEEE 88th Vehicular Technology Conference

PB - Institute of Electrical and Electronics Engineers

CY - Piscataway

ER -

Abanto Leon LF, Koppelaar AGC, Heemstra de Groot SM. TDOA-based localization via stochastic gradient descent variants. In IEEE 88th Vehicular Technology Conference: VTC 2018-Fall. Piscataway: Institute of Electrical and Electronics Engineers. 2019. 8690742 https://doi.org/10.1109/VTCFall.2018.8690742