Improving system reliability in optical networks by failure localization using evolutionary optimization

Krisztian Balazs, Peter Balazs Soproni, Laszlo T. Koczy

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

1 Citation (Scopus)

Abstract

This paper proposes a novel approach for cost-effective link failure localization in optical networks in order to improve the reliability of telecommunication systems. In such failure localization problems the optical network is usually represented by a graph, where the task is to form connected edge sets, so-called monitoring trails (m-trails), in a way that the failure of a link causes the failure of such a combination of m-trails, which unambiguously identifies the failed link. Every m-trail consumes a given amount of resources (like bandwidth, detectors, amplifiers, etc.). Thus, operators of optical network may prefer a set of paths, whose paths can be established in an easy and cost-effective way, while minimizing the interference with the route of the existing demands, i.e. may maximize the revenue. In this paper, unlike most existing techniques dealing with failure localization in this context, the presently proposed method considers a predefined set of paths in the graph as m-trails. This way the task can also be formulated as a special Set Covering Problem (SCP), whose general form is a frequently used formulation in a certain type of operations research problems (e.g. resource assignment). Since for the SCP task evolutionary algorithms, like Ant Colony Optimization (ACO), has been successfully applied in the operations research field, in this work the failure localization task is solved by using ACO on the SCP formulation of the described covering problem, which is a rather unique combination of approaches of different fields (telecommunication, operations research and evolutionary computation) placing our investigation in the multi-field scope of complex systems.

Original languageEnglish
Title of host publicationSysCon 2013 - 7th Annual IEEE International Systems Conference, Proceedings
Pages394-399
Number of pages6
DOIs
Publication statusPublished - 2013
Event7th Annual IEEE International Systems Conference, SysCon 2013 - Orlando, FL, United States
Duration: Apr 15 2013Apr 18 2013

Other

Other7th Annual IEEE International Systems Conference, SysCon 2013
CountryUnited States
CityOrlando, FL
Period4/15/134/18/13

Fingerprint

Fiber optic networks
Operations research
Monitoring
Ant colony optimization
Evolutionary algorithms
Telecommunication systems
Telecommunication
Large scale systems
Costs
Detectors
Bandwidth

Keywords

  • Ant Colony Optimization
  • Evolutionary algorithms
  • Failure localization
  • Optical networks
  • Systems reliability

ASJC Scopus subject areas

  • Control and Systems Engineering

Cite this

Balazs, K., Soproni, P. B., & Koczy, L. T. (2013). Improving system reliability in optical networks by failure localization using evolutionary optimization. In SysCon 2013 - 7th Annual IEEE International Systems Conference, Proceedings (pp. 394-399). [6549912] https://doi.org/10.1109/SysCon.2013.6549912

Improving system reliability in optical networks by failure localization using evolutionary optimization. / Balazs, Krisztian; Soproni, Peter Balazs; Koczy, Laszlo T.

SysCon 2013 - 7th Annual IEEE International Systems Conference, Proceedings. 2013. p. 394-399 6549912.

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

Balazs, K, Soproni, PB & Koczy, LT 2013, Improving system reliability in optical networks by failure localization using evolutionary optimization. in SysCon 2013 - 7th Annual IEEE International Systems Conference, Proceedings., 6549912, pp. 394-399, 7th Annual IEEE International Systems Conference, SysCon 2013, Orlando, FL, United States, 4/15/13. https://doi.org/10.1109/SysCon.2013.6549912
Balazs K, Soproni PB, Koczy LT. Improving system reliability in optical networks by failure localization using evolutionary optimization. In SysCon 2013 - 7th Annual IEEE International Systems Conference, Proceedings. 2013. p. 394-399. 6549912 https://doi.org/10.1109/SysCon.2013.6549912
Balazs, Krisztian ; Soproni, Peter Balazs ; Koczy, Laszlo T. / Improving system reliability in optical networks by failure localization using evolutionary optimization. SysCon 2013 - 7th Annual IEEE International Systems Conference, Proceedings. 2013. pp. 394-399
@inproceedings{22c8d61a572b452a89610cf614119390,
title = "Improving system reliability in optical networks by failure localization using evolutionary optimization",
abstract = "This paper proposes a novel approach for cost-effective link failure localization in optical networks in order to improve the reliability of telecommunication systems. In such failure localization problems the optical network is usually represented by a graph, where the task is to form connected edge sets, so-called monitoring trails (m-trails), in a way that the failure of a link causes the failure of such a combination of m-trails, which unambiguously identifies the failed link. Every m-trail consumes a given amount of resources (like bandwidth, detectors, amplifiers, etc.). Thus, operators of optical network may prefer a set of paths, whose paths can be established in an easy and cost-effective way, while minimizing the interference with the route of the existing demands, i.e. may maximize the revenue. In this paper, unlike most existing techniques dealing with failure localization in this context, the presently proposed method considers a predefined set of paths in the graph as m-trails. This way the task can also be formulated as a special Set Covering Problem (SCP), whose general form is a frequently used formulation in a certain type of operations research problems (e.g. resource assignment). Since for the SCP task evolutionary algorithms, like Ant Colony Optimization (ACO), has been successfully applied in the operations research field, in this work the failure localization task is solved by using ACO on the SCP formulation of the described covering problem, which is a rather unique combination of approaches of different fields (telecommunication, operations research and evolutionary computation) placing our investigation in the multi-field scope of complex systems.",
keywords = "Ant Colony Optimization, Evolutionary algorithms, Failure localization, Optical networks, Systems reliability",
author = "Krisztian Balazs and Soproni, {Peter Balazs} and Koczy, {Laszlo T.}",
year = "2013",
doi = "10.1109/SysCon.2013.6549912",
language = "English",
isbn = "9781467331067",
pages = "394--399",
booktitle = "SysCon 2013 - 7th Annual IEEE International Systems Conference, Proceedings",

}

TY - GEN

T1 - Improving system reliability in optical networks by failure localization using evolutionary optimization

AU - Balazs, Krisztian

AU - Soproni, Peter Balazs

AU - Koczy, Laszlo T.

PY - 2013

Y1 - 2013

N2 - This paper proposes a novel approach for cost-effective link failure localization in optical networks in order to improve the reliability of telecommunication systems. In such failure localization problems the optical network is usually represented by a graph, where the task is to form connected edge sets, so-called monitoring trails (m-trails), in a way that the failure of a link causes the failure of such a combination of m-trails, which unambiguously identifies the failed link. Every m-trail consumes a given amount of resources (like bandwidth, detectors, amplifiers, etc.). Thus, operators of optical network may prefer a set of paths, whose paths can be established in an easy and cost-effective way, while minimizing the interference with the route of the existing demands, i.e. may maximize the revenue. In this paper, unlike most existing techniques dealing with failure localization in this context, the presently proposed method considers a predefined set of paths in the graph as m-trails. This way the task can also be formulated as a special Set Covering Problem (SCP), whose general form is a frequently used formulation in a certain type of operations research problems (e.g. resource assignment). Since for the SCP task evolutionary algorithms, like Ant Colony Optimization (ACO), has been successfully applied in the operations research field, in this work the failure localization task is solved by using ACO on the SCP formulation of the described covering problem, which is a rather unique combination of approaches of different fields (telecommunication, operations research and evolutionary computation) placing our investigation in the multi-field scope of complex systems.

AB - This paper proposes a novel approach for cost-effective link failure localization in optical networks in order to improve the reliability of telecommunication systems. In such failure localization problems the optical network is usually represented by a graph, where the task is to form connected edge sets, so-called monitoring trails (m-trails), in a way that the failure of a link causes the failure of such a combination of m-trails, which unambiguously identifies the failed link. Every m-trail consumes a given amount of resources (like bandwidth, detectors, amplifiers, etc.). Thus, operators of optical network may prefer a set of paths, whose paths can be established in an easy and cost-effective way, while minimizing the interference with the route of the existing demands, i.e. may maximize the revenue. In this paper, unlike most existing techniques dealing with failure localization in this context, the presently proposed method considers a predefined set of paths in the graph as m-trails. This way the task can also be formulated as a special Set Covering Problem (SCP), whose general form is a frequently used formulation in a certain type of operations research problems (e.g. resource assignment). Since for the SCP task evolutionary algorithms, like Ant Colony Optimization (ACO), has been successfully applied in the operations research field, in this work the failure localization task is solved by using ACO on the SCP formulation of the described covering problem, which is a rather unique combination of approaches of different fields (telecommunication, operations research and evolutionary computation) placing our investigation in the multi-field scope of complex systems.

KW - Ant Colony Optimization

KW - Evolutionary algorithms

KW - Failure localization

KW - Optical networks

KW - Systems reliability

UR - http://www.scopus.com/inward/record.url?scp=84882936656&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84882936656&partnerID=8YFLogxK

U2 - 10.1109/SysCon.2013.6549912

DO - 10.1109/SysCon.2013.6549912

M3 - Conference contribution

AN - SCOPUS:84882936656

SN - 9781467331067

SP - 394

EP - 399

BT - SysCon 2013 - 7th Annual IEEE International Systems Conference, Proceedings

ER -