Toward a robust sparse data representation for wireless sensor networks

Mohammad Abu Alsheikh, Shaowei Lin, Hwee Pink Tan, Dusit Niyato

Research output: A Conference proceeding or a Chapter in BookConference contribution

8 Citations (Scopus)

Abstract

Compressive sensing has been successfully used for optimized operations in wireless sensor networks. However, raw data collected by sensors may be neither originally sparse nor easily transformed into a sparse data representation. This paper addresses the problem of transforming source data collected by sensor nodes into a sparse representation with a few nonzero elements. Our contributions that address three major issues include: 1) an effective method that extracts population sparsity of the data, 2) a sparsity ratio guarantee scheme, and 3) a customized learning algorithm of the sparsifying dictionary. We introduce an unsupervised neural network to extract an intrinsic sparse coding of the data. The sparse codes are generated at the activation of the hidden layer using a sparsity nomination constraint and a shrinking mechanism. Our analysis using real data samples shows that the proposed method outperforms conventional sparsity-inducing methods.

Original languageEnglish
Title of host publicationProceedings - Conference on Local Computer Networks, LCN
Place of PublicationClearwater Beach, US
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages117-124
Number of pages8
ISBN (Electronic)9781467367707
ISBN (Print)9781467367707
DOIs
Publication statusPublished - 26 Oct 2015
Externally publishedYes
Event2015 IEEE 40th Conference on Local Computer Networks, LCN 2015 - Clearwater Beach, United States
Duration: 26 Oct 201529 Oct 2015

Publication series

NameProceedings - Conference on Local Computer Networks, LCN
Volume26-29-October-2015

Conference

Conference2015 IEEE 40th Conference on Local Computer Networks, LCN 2015
CountryUnited States
CityClearwater Beach
Period26/10/1529/10/15

Fingerprint

Glossaries
Sensor nodes
Learning algorithms
Wireless sensor networks
Chemical activation
Neural networks
Sensors

Cite this

Abu Alsheikh, M., Lin, S., Tan, H. P., & Niyato, D. (2015). Toward a robust sparse data representation for wireless sensor networks. In Proceedings - Conference on Local Computer Networks, LCN (pp. 117-124). [7366290] (Proceedings - Conference on Local Computer Networks, LCN; Vol. 26-29-October-2015). Clearwater Beach, US: IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/LCN.2015.7366290
Abu Alsheikh, Mohammad ; Lin, Shaowei ; Tan, Hwee Pink ; Niyato, Dusit. / Toward a robust sparse data representation for wireless sensor networks. Proceedings - Conference on Local Computer Networks, LCN. Clearwater Beach, US : IEEE, Institute of Electrical and Electronics Engineers, 2015. pp. 117-124 (Proceedings - Conference on Local Computer Networks, LCN).
@inproceedings{d08cdbaea3e549d18384a4c3f55f0e54,
title = "Toward a robust sparse data representation for wireless sensor networks",
abstract = "Compressive sensing has been successfully used for optimized operations in wireless sensor networks. However, raw data collected by sensors may be neither originally sparse nor easily transformed into a sparse data representation. This paper addresses the problem of transforming source data collected by sensor nodes into a sparse representation with a few nonzero elements. Our contributions that address three major issues include: 1) an effective method that extracts population sparsity of the data, 2) a sparsity ratio guarantee scheme, and 3) a customized learning algorithm of the sparsifying dictionary. We introduce an unsupervised neural network to extract an intrinsic sparse coding of the data. The sparse codes are generated at the activation of the hidden layer using a sparsity nomination constraint and a shrinking mechanism. Our analysis using real data samples shows that the proposed method outperforms conventional sparsity-inducing methods.",
keywords = "compressive sensing, sparse autoencoders, Sparse coding, wireless sensor networks",
author = "{Abu Alsheikh}, Mohammad and Shaowei Lin and Tan, {Hwee Pink} and Dusit Niyato",
year = "2015",
month = "10",
day = "26",
doi = "10.1109/LCN.2015.7366290",
language = "English",
isbn = "9781467367707",
series = "Proceedings - Conference on Local Computer Networks, LCN",
publisher = "IEEE, Institute of Electrical and Electronics Engineers",
pages = "117--124",
booktitle = "Proceedings - Conference on Local Computer Networks, LCN",
address = "United States",

}

Abu Alsheikh, M, Lin, S, Tan, HP & Niyato, D 2015, Toward a robust sparse data representation for wireless sensor networks. in Proceedings - Conference on Local Computer Networks, LCN., 7366290, Proceedings - Conference on Local Computer Networks, LCN, vol. 26-29-October-2015, IEEE, Institute of Electrical and Electronics Engineers, Clearwater Beach, US, pp. 117-124, 2015 IEEE 40th Conference on Local Computer Networks, LCN 2015, Clearwater Beach, United States, 26/10/15. https://doi.org/10.1109/LCN.2015.7366290

Toward a robust sparse data representation for wireless sensor networks. / Abu Alsheikh, Mohammad; Lin, Shaowei; Tan, Hwee Pink; Niyato, Dusit.

Proceedings - Conference on Local Computer Networks, LCN. Clearwater Beach, US : IEEE, Institute of Electrical and Electronics Engineers, 2015. p. 117-124 7366290 (Proceedings - Conference on Local Computer Networks, LCN; Vol. 26-29-October-2015).

Research output: A Conference proceeding or a Chapter in BookConference contribution

TY - GEN

T1 - Toward a robust sparse data representation for wireless sensor networks

AU - Abu Alsheikh, Mohammad

AU - Lin, Shaowei

AU - Tan, Hwee Pink

AU - Niyato, Dusit

PY - 2015/10/26

Y1 - 2015/10/26

N2 - Compressive sensing has been successfully used for optimized operations in wireless sensor networks. However, raw data collected by sensors may be neither originally sparse nor easily transformed into a sparse data representation. This paper addresses the problem of transforming source data collected by sensor nodes into a sparse representation with a few nonzero elements. Our contributions that address three major issues include: 1) an effective method that extracts population sparsity of the data, 2) a sparsity ratio guarantee scheme, and 3) a customized learning algorithm of the sparsifying dictionary. We introduce an unsupervised neural network to extract an intrinsic sparse coding of the data. The sparse codes are generated at the activation of the hidden layer using a sparsity nomination constraint and a shrinking mechanism. Our analysis using real data samples shows that the proposed method outperforms conventional sparsity-inducing methods.

AB - Compressive sensing has been successfully used for optimized operations in wireless sensor networks. However, raw data collected by sensors may be neither originally sparse nor easily transformed into a sparse data representation. This paper addresses the problem of transforming source data collected by sensor nodes into a sparse representation with a few nonzero elements. Our contributions that address three major issues include: 1) an effective method that extracts population sparsity of the data, 2) a sparsity ratio guarantee scheme, and 3) a customized learning algorithm of the sparsifying dictionary. We introduce an unsupervised neural network to extract an intrinsic sparse coding of the data. The sparse codes are generated at the activation of the hidden layer using a sparsity nomination constraint and a shrinking mechanism. Our analysis using real data samples shows that the proposed method outperforms conventional sparsity-inducing methods.

KW - compressive sensing

KW - sparse autoencoders

KW - Sparse coding

KW - wireless sensor networks

UR - http://www.scopus.com/inward/record.url?scp=84973926638&partnerID=8YFLogxK

U2 - 10.1109/LCN.2015.7366290

DO - 10.1109/LCN.2015.7366290

M3 - Conference contribution

SN - 9781467367707

T3 - Proceedings - Conference on Local Computer Networks, LCN

SP - 117

EP - 124

BT - Proceedings - Conference on Local Computer Networks, LCN

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - Clearwater Beach, US

ER -

Abu Alsheikh M, Lin S, Tan HP, Niyato D. Toward a robust sparse data representation for wireless sensor networks. In Proceedings - Conference on Local Computer Networks, LCN. Clearwater Beach, US: IEEE, Institute of Electrical and Electronics Engineers. 2015. p. 117-124. 7366290. (Proceedings - Conference on Local Computer Networks, LCN). https://doi.org/10.1109/LCN.2015.7366290