Trade-offs in the distribution of neural networks in a wireless sensor network

M.J. Holenderski; J.J. Lukkien; C.K. Tham

doi:10.1109/ISSNIP.2005.1595589

Trade-offs in the distribution of neural networks in a wireless sensor network

M.J. Holenderski, J.J. Lukkien, C.K. Tham

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

3 Citations (Scopus)

Abstract

This article investigates the tradeoff between communication and memory usage in different methods of distributing neural networks in a Wireless Sensor Network. A structural approach is presented, categorized in two dimensions: horizontal and vertical decomposition. Horizontal decomposition turns out to be more attractive, due to high reuse of data present at the processor node. General properties of an alternative sematic approach are suggested theoretically allowing to dramatically increase efficiency.

Original language	English
Title of host publication	Intelligent Sensors, Sensor Networks and Information Processing Proceedings of the 2nd International Conference, ISSNIP'05, Melbourne, Australia, December 5-8, 2005
Publisher	Institute of Electrical and Electronics Engineers
Pages	259-264
ISBN (Print)	0-7803-9399-6
DOIs	https://doi.org/10.1109/ISSNIP.2005.1595589
Publication status	Published - 2005

Access to Document

10.1109/ISSNIP.2005.1595589

Cite this

Holenderski, M. J., Lukkien, J. J., & Tham, C. K. (2005). Trade-offs in the distribution of neural networks in a wireless sensor network. In Intelligent Sensors, Sensor Networks and Information Processing Proceedings of the 2nd International Conference, ISSNIP'05, Melbourne, Australia, December 5-8, 2005 (pp. 259-264). Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/ISSNIP.2005.1595589

@inproceedings{49e9f19aa1f64546958f2ef053ac6fae,

title = "Trade-offs in the distribution of neural networks in a wireless sensor network",

abstract = "This article investigates the tradeoff between communication and memory usage in different methods of distributing neural networks in a Wireless Sensor Network. A structural approach is presented, categorized in two dimensions: horizontal and vertical decomposition. Horizontal decomposition turns out to be more attractive, due to high reuse of data present at the processor node. General properties of an alternative sematic approach are suggested theoretically allowing to dramatically increase efficiency.",

author = "M.J. Holenderski and J.J. Lukkien and C.K. Tham",

year = "2005",

doi = "10.1109/ISSNIP.2005.1595589",

language = "English",

isbn = "0-7803-9399-6",

pages = "259--264",

booktitle = "Intelligent Sensors, Sensor Networks and Information Processing Proceedings of the 2nd International Conference, ISSNIP'05, Melbourne, Australia, December 5-8, 2005",

publisher = "Institute of Electrical and Electronics Engineers",

address = "United States",

}

Holenderski, MJ , Lukkien, JJ & Tham, CK 2005, Trade-offs in the distribution of neural networks in a wireless sensor network. in Intelligent Sensors, Sensor Networks and Information Processing Proceedings of the 2nd International Conference, ISSNIP'05, Melbourne, Australia, December 5-8, 2005. Institute of Electrical and Electronics Engineers, pp. 259-264. https://doi.org/10.1109/ISSNIP.2005.1595589

Trade-offs in the distribution of neural networks in a wireless sensor network. / Holenderski, M.J.; Lukkien, J.J.; Tham, C.K.

Intelligent Sensors, Sensor Networks and Information Processing Proceedings of the 2nd International Conference, ISSNIP'05, Melbourne, Australia, December 5-8, 2005. Institute of Electrical and Electronics Engineers, 2005. p. 259-264.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Trade-offs in the distribution of neural networks in a wireless sensor network

AU - Holenderski, M.J.

AU - Lukkien, J.J.

AU - Tham, C.K.

PY - 2005

Y1 - 2005

N2 - This article investigates the tradeoff between communication and memory usage in different methods of distributing neural networks in a Wireless Sensor Network. A structural approach is presented, categorized in two dimensions: horizontal and vertical decomposition. Horizontal decomposition turns out to be more attractive, due to high reuse of data present at the processor node. General properties of an alternative sematic approach are suggested theoretically allowing to dramatically increase efficiency.

AB - This article investigates the tradeoff between communication and memory usage in different methods of distributing neural networks in a Wireless Sensor Network. A structural approach is presented, categorized in two dimensions: horizontal and vertical decomposition. Horizontal decomposition turns out to be more attractive, due to high reuse of data present at the processor node. General properties of an alternative sematic approach are suggested theoretically allowing to dramatically increase efficiency.

U2 - 10.1109/ISSNIP.2005.1595589

DO - 10.1109/ISSNIP.2005.1595589

M3 - Conference contribution

SN - 0-7803-9399-6

SP - 259

EP - 264

BT - Intelligent Sensors, Sensor Networks and Information Processing Proceedings of the 2nd International Conference, ISSNIP'05, Melbourne, Australia, December 5-8, 2005

PB - Institute of Electrical and Electronics Engineers

ER -

Holenderski MJ , Lukkien JJ, Tham CK. Trade-offs in the distribution of neural networks in a wireless sensor network. In Intelligent Sensors, Sensor Networks and Information Processing Proceedings of the 2nd International Conference, ISSNIP'05, Melbourne, Australia, December 5-8, 2005. Institute of Electrical and Electronics Engineers. 2005. p. 259-264 https://doi.org/10.1109/ISSNIP.2005.1595589

Trade-offs in the distribution of neural networks in a wireless sensor network

Abstract

Access to Document

Fingerprint

Cite this