The Power-Oja method for decentralized subspace estimation/tracking

Sissi Xiaoxiao Wu, Hoi To Wai, Anna Scaglione, Neil A. Jacklin

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

This work proposes a decentralized and adaptive subspace estimation method, called the Power-Oja (P-Oja) method. Existing decentralized subspace tracking algorithms have slow convergence rate or are unable to adapt to time varying statistics. To resolve these issues, the P-Oja method is developed by combining the power method with Oja's learning rule. Our key innovation lies on the design of a modified objective function with enhanced spectral gap property. This allows the P-Oja method to track the principal subspace more quickly with a finite number of samples. Interestingly, the resulting method coincides with the conventional Oja's learning rule in some special cases. To enable decentralized signal processing, we further demonstrate that the proposed method can be implemented by using a gossip algorithm. Our simulation results show that the proposed P-Oja outperforms the conventional Oja's method in terms of estimation accuracy, and the power method in terms of tracking performance. The effect of the communication graph on the tracking performance is also studied.

Original languageEnglish (US)
Title of host publication2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages3524-3528
Number of pages5
ISBN (Electronic)9781509041176
DOIs
StatePublished - Jun 16 2017
Event2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - New Orleans, United States
Duration: Mar 5 2017Mar 9 2017

Other

Other2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017
CountryUnited States
CityNew Orleans
Period3/5/173/9/17

Fingerprint

Signal processing
Innovation
Statistics
Communication

Keywords

  • gossip algorithm
  • Massive arrays
  • spectrum sensing
  • subspace estimation

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Wu, S. X., Wai, H. T., Scaglione, A., & Jacklin, N. A. (2017). The Power-Oja method for decentralized subspace estimation/tracking. In 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings (pp. 3524-3528). [7952812] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICASSP.2017.7952812

The Power-Oja method for decentralized subspace estimation/tracking. / Wu, Sissi Xiaoxiao; Wai, Hoi To; Scaglione, Anna; Jacklin, Neil A.

2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2017. p. 3524-3528 7952812.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Wu, SX, Wai, HT, Scaglione, A & Jacklin, NA 2017, The Power-Oja method for decentralized subspace estimation/tracking. in 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings., 7952812, Institute of Electrical and Electronics Engineers Inc., pp. 3524-3528, 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017, New Orleans, United States, 3/5/17. https://doi.org/10.1109/ICASSP.2017.7952812
Wu SX, Wai HT, Scaglione A, Jacklin NA. The Power-Oja method for decentralized subspace estimation/tracking. In 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2017. p. 3524-3528. 7952812 https://doi.org/10.1109/ICASSP.2017.7952812
Wu, Sissi Xiaoxiao ; Wai, Hoi To ; Scaglione, Anna ; Jacklin, Neil A. / The Power-Oja method for decentralized subspace estimation/tracking. 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 3524-3528
@inproceedings{00e686b7f8844d82af496a50deed9b65,
title = "The Power-Oja method for decentralized subspace estimation/tracking",
abstract = "This work proposes a decentralized and adaptive subspace estimation method, called the Power-Oja (P-Oja) method. Existing decentralized subspace tracking algorithms have slow convergence rate or are unable to adapt to time varying statistics. To resolve these issues, the P-Oja method is developed by combining the power method with Oja's learning rule. Our key innovation lies on the design of a modified objective function with enhanced spectral gap property. This allows the P-Oja method to track the principal subspace more quickly with a finite number of samples. Interestingly, the resulting method coincides with the conventional Oja's learning rule in some special cases. To enable decentralized signal processing, we further demonstrate that the proposed method can be implemented by using a gossip algorithm. Our simulation results show that the proposed P-Oja outperforms the conventional Oja's method in terms of estimation accuracy, and the power method in terms of tracking performance. The effect of the communication graph on the tracking performance is also studied.",
keywords = "gossip algorithm, Massive arrays, spectrum sensing, subspace estimation",
author = "Wu, {Sissi Xiaoxiao} and Wai, {Hoi To} and Anna Scaglione and Jacklin, {Neil A.}",
year = "2017",
month = "6",
day = "16",
doi = "10.1109/ICASSP.2017.7952812",
language = "English (US)",
pages = "3524--3528",
booktitle = "2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",

}

TY - GEN

T1 - The Power-Oja method for decentralized subspace estimation/tracking

AU - Wu, Sissi Xiaoxiao

AU - Wai, Hoi To

AU - Scaglione, Anna

AU - Jacklin, Neil A.

PY - 2017/6/16

Y1 - 2017/6/16

N2 - This work proposes a decentralized and adaptive subspace estimation method, called the Power-Oja (P-Oja) method. Existing decentralized subspace tracking algorithms have slow convergence rate or are unable to adapt to time varying statistics. To resolve these issues, the P-Oja method is developed by combining the power method with Oja's learning rule. Our key innovation lies on the design of a modified objective function with enhanced spectral gap property. This allows the P-Oja method to track the principal subspace more quickly with a finite number of samples. Interestingly, the resulting method coincides with the conventional Oja's learning rule in some special cases. To enable decentralized signal processing, we further demonstrate that the proposed method can be implemented by using a gossip algorithm. Our simulation results show that the proposed P-Oja outperforms the conventional Oja's method in terms of estimation accuracy, and the power method in terms of tracking performance. The effect of the communication graph on the tracking performance is also studied.

AB - This work proposes a decentralized and adaptive subspace estimation method, called the Power-Oja (P-Oja) method. Existing decentralized subspace tracking algorithms have slow convergence rate or are unable to adapt to time varying statistics. To resolve these issues, the P-Oja method is developed by combining the power method with Oja's learning rule. Our key innovation lies on the design of a modified objective function with enhanced spectral gap property. This allows the P-Oja method to track the principal subspace more quickly with a finite number of samples. Interestingly, the resulting method coincides with the conventional Oja's learning rule in some special cases. To enable decentralized signal processing, we further demonstrate that the proposed method can be implemented by using a gossip algorithm. Our simulation results show that the proposed P-Oja outperforms the conventional Oja's method in terms of estimation accuracy, and the power method in terms of tracking performance. The effect of the communication graph on the tracking performance is also studied.

KW - gossip algorithm

KW - Massive arrays

KW - spectrum sensing

KW - subspace estimation

UR - http://www.scopus.com/inward/record.url?scp=85023768984&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85023768984&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2017.7952812

DO - 10.1109/ICASSP.2017.7952812

M3 - Conference contribution

AN - SCOPUS:85023768984

SP - 3524

EP - 3528

BT - 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -