Adaptive In-network processing for bandwidth and energy constrained mission-oriented multi-hop wireless networks

Sharanya Eswaran, Matthew Johnson, Archan Misra, Thomas F. La Porta

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

10 Citations (Scopus)

Abstract

In-network processing, involving operations such as filtering, compression and fusion, is widely used in sensor networks to reduce the communication overhead. In many tactical and stream-oriented wireless network applications, both link bandwidth and node energy are critically constrained resources and in-network processing itself imposes non-negligible computing cost. In this work, we have developed a unified and distributed closed-loop control framework that computes both a) the optimal level of sensor stream compression performed by a forwarding node, and b) the best set of nodes where the stream processing operators should be deployed. Our framework extends the Network Utility Maximization (NUM) paradigm, where resource sharing among competing applications is modeled as a form of distributed utility maximization. We also show how our model can be adapted to more realistic cases, where in-network compression may be varied only discretely, and where a fusion operation cannot be fractionally distributed across multiple nodes.

Original languageEnglish (US)
Title of host publicationDistributed Computing in Sensor Systems - 5th IEEE International Conference, DCOSS 2009, Proceedings
Pages87-102
Number of pages16
DOIs
StatePublished - Aug 20 2009
Event5th IEEE International Conference on Distributed Computing in Sensor Systems, DCOSS 2009 - Marina del Rey, CA, United States
Duration: Jun 8 2009Jun 10 2009

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume5516 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other5th IEEE International Conference on Distributed Computing in Sensor Systems, DCOSS 2009
CountryUnited States
CityMarina del Rey, CA
Period6/8/096/10/09

Fingerprint

Multi-hop Wireless Networks
Wireless networks
Bandwidth
Utility Maximization
Compression
Fusion reactions
Vertex of a graph
Processing
Energy
Fusion
Stream Processing
Sensor networks
Telecommunication links
Resource Sharing
Closed-loop Control
Distributed Control
Sensor Networks
Wireless Networks
Filtering
Communication

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Eswaran, S., Johnson, M., Misra, A., & La Porta, T. F. (2009). Adaptive In-network processing for bandwidth and energy constrained mission-oriented multi-hop wireless networks. In Distributed Computing in Sensor Systems - 5th IEEE International Conference, DCOSS 2009, Proceedings (pp. 87-102). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5516 LNCS). https://doi.org/10.1007/978-3-642-02085-8_7
Eswaran, Sharanya ; Johnson, Matthew ; Misra, Archan ; La Porta, Thomas F. / Adaptive In-network processing for bandwidth and energy constrained mission-oriented multi-hop wireless networks. Distributed Computing in Sensor Systems - 5th IEEE International Conference, DCOSS 2009, Proceedings. 2009. pp. 87-102 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{ecd0c8ff7a094a3fa6a11db790d937fa,
title = "Adaptive In-network processing for bandwidth and energy constrained mission-oriented multi-hop wireless networks",
abstract = "In-network processing, involving operations such as filtering, compression and fusion, is widely used in sensor networks to reduce the communication overhead. In many tactical and stream-oriented wireless network applications, both link bandwidth and node energy are critically constrained resources and in-network processing itself imposes non-negligible computing cost. In this work, we have developed a unified and distributed closed-loop control framework that computes both a) the optimal level of sensor stream compression performed by a forwarding node, and b) the best set of nodes where the stream processing operators should be deployed. Our framework extends the Network Utility Maximization (NUM) paradigm, where resource sharing among competing applications is modeled as a form of distributed utility maximization. We also show how our model can be adapted to more realistic cases, where in-network compression may be varied only discretely, and where a fusion operation cannot be fractionally distributed across multiple nodes.",
author = "Sharanya Eswaran and Matthew Johnson and Archan Misra and {La Porta}, {Thomas F.}",
year = "2009",
month = "8",
day = "20",
doi = "10.1007/978-3-642-02085-8_7",
language = "English (US)",
isbn = "3642020844",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
pages = "87--102",
booktitle = "Distributed Computing in Sensor Systems - 5th IEEE International Conference, DCOSS 2009, Proceedings",

}

Eswaran, S, Johnson, M, Misra, A & La Porta, TF 2009, Adaptive In-network processing for bandwidth and energy constrained mission-oriented multi-hop wireless networks. in Distributed Computing in Sensor Systems - 5th IEEE International Conference, DCOSS 2009, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 5516 LNCS, pp. 87-102, 5th IEEE International Conference on Distributed Computing in Sensor Systems, DCOSS 2009, Marina del Rey, CA, United States, 6/8/09. https://doi.org/10.1007/978-3-642-02085-8_7

Adaptive In-network processing for bandwidth and energy constrained mission-oriented multi-hop wireless networks. / Eswaran, Sharanya; Johnson, Matthew; Misra, Archan; La Porta, Thomas F.

Distributed Computing in Sensor Systems - 5th IEEE International Conference, DCOSS 2009, Proceedings. 2009. p. 87-102 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5516 LNCS).

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

TY - GEN

T1 - Adaptive In-network processing for bandwidth and energy constrained mission-oriented multi-hop wireless networks

AU - Eswaran, Sharanya

AU - Johnson, Matthew

AU - Misra, Archan

AU - La Porta, Thomas F.

PY - 2009/8/20

Y1 - 2009/8/20

N2 - In-network processing, involving operations such as filtering, compression and fusion, is widely used in sensor networks to reduce the communication overhead. In many tactical and stream-oriented wireless network applications, both link bandwidth and node energy are critically constrained resources and in-network processing itself imposes non-negligible computing cost. In this work, we have developed a unified and distributed closed-loop control framework that computes both a) the optimal level of sensor stream compression performed by a forwarding node, and b) the best set of nodes where the stream processing operators should be deployed. Our framework extends the Network Utility Maximization (NUM) paradigm, where resource sharing among competing applications is modeled as a form of distributed utility maximization. We also show how our model can be adapted to more realistic cases, where in-network compression may be varied only discretely, and where a fusion operation cannot be fractionally distributed across multiple nodes.

AB - In-network processing, involving operations such as filtering, compression and fusion, is widely used in sensor networks to reduce the communication overhead. In many tactical and stream-oriented wireless network applications, both link bandwidth and node energy are critically constrained resources and in-network processing itself imposes non-negligible computing cost. In this work, we have developed a unified and distributed closed-loop control framework that computes both a) the optimal level of sensor stream compression performed by a forwarding node, and b) the best set of nodes where the stream processing operators should be deployed. Our framework extends the Network Utility Maximization (NUM) paradigm, where resource sharing among competing applications is modeled as a form of distributed utility maximization. We also show how our model can be adapted to more realistic cases, where in-network compression may be varied only discretely, and where a fusion operation cannot be fractionally distributed across multiple nodes.

UR - http://www.scopus.com/inward/record.url?scp=68749086471&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=68749086471&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-02085-8_7

DO - 10.1007/978-3-642-02085-8_7

M3 - Conference contribution

SN - 3642020844

SN - 9783642020841

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 87

EP - 102

BT - Distributed Computing in Sensor Systems - 5th IEEE International Conference, DCOSS 2009, Proceedings

ER -

Eswaran S, Johnson M, Misra A, La Porta TF. Adaptive In-network processing for bandwidth and energy constrained mission-oriented multi-hop wireless networks. In Distributed Computing in Sensor Systems - 5th IEEE International Conference, DCOSS 2009, Proceedings. 2009. p. 87-102. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-642-02085-8_7