BARD: A Novel Application of Bayesian Reasoning for Proactive Network Management

Abul Bashar, Gerard Parr, Sally McClean, Bryan Scotney, D Nauck

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

Abstract

In the context of next generation networks (NGN), there is a critical need to address the challenges facing the network management functions, to administer the converged infrastructure and services. One of the challenges is to model the unknown and probabilistic dependency relationships among the diverse network elements and adapt this model to capture the real-time network behavior. This paper proposes BARD (BAyesian Reasoner and Decision-maker), a proactive system to enhance the network performance management functions by use of machine learning technique called Bayesian Belief Networks (BBN). It exploits the predictive and diagnostic reasoning features of BBN to make accurate decisions for effective management. A case study is presented to demonstrate the application of this technique to the problem of Call Admission Control in a typical communication network. The simulation results provide the proof that BARD is an effective and feasible solution when applied for network management tasks, especially Quality-of-Service (QoS) management.
LanguageEnglish
Title of host publicationUnknown Host Publication
Number of pages6
Publication statusPublished - Jun 2009
Event10th PostGraduate Conference on Telecoms, Networking and Broadcasting - Liverpool
Duration: 1 Jun 2009 → …

Conference

Conference10th PostGraduate Conference on Telecoms, Networking and Broadcasting
Period1/06/09 → …

Fingerprint

Network management
Bayesian networks
Next generation networks
Congestion control (communication)
Network performance
Telecommunication networks
Learning systems
Quality of service

Keywords

  • Network Management
  • Next Generation Networks (NGN)
  • Bayesian Belief Networks (BBN)
  • Machine Learning.

Cite this

Bashar, A., Parr, G., McClean, S., Scotney, B., & Nauck, D. (2009). BARD: A Novel Application of Bayesian Reasoning for Proactive Network Management. In Unknown Host Publication
@inproceedings{1f4d7781b1c24ed4b09e71534d04a77b,
title = "BARD: A Novel Application of Bayesian Reasoning for Proactive Network Management",
abstract = "In the context of next generation networks (NGN), there is a critical need to address the challenges facing the network management functions, to administer the converged infrastructure and services. One of the challenges is to model the unknown and probabilistic dependency relationships among the diverse network elements and adapt this model to capture the real-time network behavior. This paper proposes BARD (BAyesian Reasoner and Decision-maker), a proactive system to enhance the network performance management functions by use of machine learning technique called Bayesian Belief Networks (BBN). It exploits the predictive and diagnostic reasoning features of BBN to make accurate decisions for effective management. A case study is presented to demonstrate the application of this technique to the problem of Call Admission Control in a typical communication network. The simulation results provide the proof that BARD is an effective and feasible solution when applied for network management tasks, especially Quality-of-Service (QoS) management.",
keywords = "Network Management, Next Generation Networks (NGN), Bayesian Belief Networks (BBN), Machine Learning.",
author = "Abul Bashar and Gerard Parr and Sally McClean and Bryan Scotney and D Nauck",
note = "Reference text: [1] General overview of NGN, ITU-T Recommendation Y.2001, Dec 2004. [2] Aiko Pras et al, “Key research challenges in network management,” IEEE Communications Magazine, Oct 2007, pp. 104-110. [3] J. Case, M. Fedor, M. Schoffstall, J. Davin, “Simple Network Management Protocol,” RFC 1157, IETF, May 1990. [4] Open systems interconnection (OSI) Common Management Information Protocol: specification, ITU-T Recommendation X.711, Oct 1997. [5] TMN Management Functions, ITU-T Recommendation M.3400, April 1997. [6] P.G. Kulkarni, S. I. McClean, G. P. Parr, M. M. Black, “Deploying MIB data mining for proactive network management,” Proc 3rd Intl. IEEE Conference on Intelligent Systems, Sep 2006, pp. 506-511. [7] S. Sohail, A. Khanum, “Simplifying network management with fuzzy logic,” IEEE Itnl. Conf. on Communications, May 2008, pp. 195-201. [8] C. S. Hood, C. Ji , “Proactive network fault detection,” IEEE Transactions on Reliability, Sep 1997, pp. 333-341. [9] J. Ding, B. Kramer et al, “Predictive fault management in the dynamic environment of IP network,” in Proc IEEE International Workshop on IP Operations and Management, Oct 2004, pp. 233-239. [10] E. U. Ekaette, B. H. Far, “A framework for distributed fault management using intelligent software agents,” Proc IEEE Canadian Conference on Electrical and Computer Engineering 2003, Vol 2, May 2003, pp. 797-800. [11] P.G. Kulkarni, S.I. McClean, G.P. Parr, M.M. Black, “Proactive predictive queue management for improved QoS in IP networks,” Proc of IEEE ICN/ICONS/MCL, Apr 2006, pp. 2-7. [12] T. C.-K. Hui, Chen-Khong Thanm , “Adaptive provisioning of differentiated services networks based on reinforcement learning ,” IEEE Transactions on Systems, Man, and Cybernetics, Vol. 33, no. 4, Nov 2003, pp. 492-501. [13] P. Spirtes, C. Glymour, R. Scheines, Causation, Prediction, and Search, MIT Press, Second edition, Jan 2001. [14] Steck, H., “Constrained-based structural learning in Bayesian networks using finite data sets”, PhD Thesis, 2001, Institut f{\"u}r der Informatik der Technischen Universit{\"a}t M{\"u}nchen. [15] S. L. Lauritzen, “The EM algorithm for graphical association models with missing data,” Computational Statistics & Data Analysis, Vol. 19, 1995, pp. 191-201. [16] D. J. Spiegelhalter & S. L. Lauritzen, “Sequential updating of conditional probabilities on directed graphical structures,” Networks, Special Issue on Influence Diagrams, Vol. 20, No. 5, Aug 1990, pp. 579-605. [17] J. Pearl, Probabilistic reasoning in intelligent systems, Morgan Kaufmann, San Mateo, CA, 1988. [18] H. G. Perros, K. M. Elsayed, “Call admission control schemes,” IEEE Communications Magazine, Nov 1996, pp. 82-91. [19] Available at: http://www.opnet.com [20] Available at: http://www.hugin.com [21] S. L. Lauritzen, D. J. Spiegelhalter, “Local computations with probabilities on graphical structures and their application to expert systems,” Journal of the Royal Statistical Society, Series B (Methodological), Vol. 50, No. 2, Jan 1988, pp. 157-224. [22] A.L. Madsen et al, “The Hugin tool for learning Bayesian networks,” LNCS, Vol 2711, Springer 2004, pp. 594-605. [23] S. F. Galan, “Belief updating in Bayesian networks by using a criterion on minimum time,” Pattern Recognition Letters, Vol 29, 2008, pp. 465-482. [24] D. Heckerman, “A tutorial on learning with Bayesian networks,” Learning in Graphical Models, M. Jordan, ed. MIT Press, Cambridge, MA, 1999.",
year = "2009",
month = "6",
language = "English",
isbn = "978-1-902560-22-9",
booktitle = "Unknown Host Publication",

}

Bashar, A, Parr, G, McClean, S, Scotney, B & Nauck, D 2009, BARD: A Novel Application of Bayesian Reasoning for Proactive Network Management. in Unknown Host Publication. 10th PostGraduate Conference on Telecoms, Networking and Broadcasting, 1/06/09.

BARD: A Novel Application of Bayesian Reasoning for Proactive Network Management. / Bashar, Abul; Parr, Gerard; McClean, Sally; Scotney, Bryan; Nauck, D.

Unknown Host Publication. 2009.

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

TY - GEN

T1 - BARD: A Novel Application of Bayesian Reasoning for Proactive Network Management

AU - Bashar, Abul

AU - Parr, Gerard

AU - McClean, Sally

AU - Scotney, Bryan

AU - Nauck, D

N1 - Reference text: [1] General overview of NGN, ITU-T Recommendation Y.2001, Dec 2004. [2] Aiko Pras et al, “Key research challenges in network management,” IEEE Communications Magazine, Oct 2007, pp. 104-110. [3] J. Case, M. Fedor, M. Schoffstall, J. Davin, “Simple Network Management Protocol,” RFC 1157, IETF, May 1990. [4] Open systems interconnection (OSI) Common Management Information Protocol: specification, ITU-T Recommendation X.711, Oct 1997. [5] TMN Management Functions, ITU-T Recommendation M.3400, April 1997. [6] P.G. Kulkarni, S. I. McClean, G. P. Parr, M. M. Black, “Deploying MIB data mining for proactive network management,” Proc 3rd Intl. IEEE Conference on Intelligent Systems, Sep 2006, pp. 506-511. [7] S. Sohail, A. Khanum, “Simplifying network management with fuzzy logic,” IEEE Itnl. Conf. on Communications, May 2008, pp. 195-201. [8] C. S. Hood, C. Ji , “Proactive network fault detection,” IEEE Transactions on Reliability, Sep 1997, pp. 333-341. [9] J. Ding, B. Kramer et al, “Predictive fault management in the dynamic environment of IP network,” in Proc IEEE International Workshop on IP Operations and Management, Oct 2004, pp. 233-239. [10] E. U. Ekaette, B. H. Far, “A framework for distributed fault management using intelligent software agents,” Proc IEEE Canadian Conference on Electrical and Computer Engineering 2003, Vol 2, May 2003, pp. 797-800. [11] P.G. Kulkarni, S.I. McClean, G.P. Parr, M.M. Black, “Proactive predictive queue management for improved QoS in IP networks,” Proc of IEEE ICN/ICONS/MCL, Apr 2006, pp. 2-7. [12] T. C.-K. Hui, Chen-Khong Thanm , “Adaptive provisioning of differentiated services networks based on reinforcement learning ,” IEEE Transactions on Systems, Man, and Cybernetics, Vol. 33, no. 4, Nov 2003, pp. 492-501. [13] P. Spirtes, C. Glymour, R. Scheines, Causation, Prediction, and Search, MIT Press, Second edition, Jan 2001. [14] Steck, H., “Constrained-based structural learning in Bayesian networks using finite data sets”, PhD Thesis, 2001, Institut für der Informatik der Technischen Universität München. [15] S. L. Lauritzen, “The EM algorithm for graphical association models with missing data,” Computational Statistics & Data Analysis, Vol. 19, 1995, pp. 191-201. [16] D. J. Spiegelhalter & S. L. Lauritzen, “Sequential updating of conditional probabilities on directed graphical structures,” Networks, Special Issue on Influence Diagrams, Vol. 20, No. 5, Aug 1990, pp. 579-605. [17] J. Pearl, Probabilistic reasoning in intelligent systems, Morgan Kaufmann, San Mateo, CA, 1988. [18] H. G. Perros, K. M. Elsayed, “Call admission control schemes,” IEEE Communications Magazine, Nov 1996, pp. 82-91. [19] Available at: http://www.opnet.com [20] Available at: http://www.hugin.com [21] S. L. Lauritzen, D. J. Spiegelhalter, “Local computations with probabilities on graphical structures and their application to expert systems,” Journal of the Royal Statistical Society, Series B (Methodological), Vol. 50, No. 2, Jan 1988, pp. 157-224. [22] A.L. Madsen et al, “The Hugin tool for learning Bayesian networks,” LNCS, Vol 2711, Springer 2004, pp. 594-605. [23] S. F. Galan, “Belief updating in Bayesian networks by using a criterion on minimum time,” Pattern Recognition Letters, Vol 29, 2008, pp. 465-482. [24] D. Heckerman, “A tutorial on learning with Bayesian networks,” Learning in Graphical Models, M. Jordan, ed. MIT Press, Cambridge, MA, 1999.

PY - 2009/6

Y1 - 2009/6

N2 - In the context of next generation networks (NGN), there is a critical need to address the challenges facing the network management functions, to administer the converged infrastructure and services. One of the challenges is to model the unknown and probabilistic dependency relationships among the diverse network elements and adapt this model to capture the real-time network behavior. This paper proposes BARD (BAyesian Reasoner and Decision-maker), a proactive system to enhance the network performance management functions by use of machine learning technique called Bayesian Belief Networks (BBN). It exploits the predictive and diagnostic reasoning features of BBN to make accurate decisions for effective management. A case study is presented to demonstrate the application of this technique to the problem of Call Admission Control in a typical communication network. The simulation results provide the proof that BARD is an effective and feasible solution when applied for network management tasks, especially Quality-of-Service (QoS) management.

AB - In the context of next generation networks (NGN), there is a critical need to address the challenges facing the network management functions, to administer the converged infrastructure and services. One of the challenges is to model the unknown and probabilistic dependency relationships among the diverse network elements and adapt this model to capture the real-time network behavior. This paper proposes BARD (BAyesian Reasoner and Decision-maker), a proactive system to enhance the network performance management functions by use of machine learning technique called Bayesian Belief Networks (BBN). It exploits the predictive and diagnostic reasoning features of BBN to make accurate decisions for effective management. A case study is presented to demonstrate the application of this technique to the problem of Call Admission Control in a typical communication network. The simulation results provide the proof that BARD is an effective and feasible solution when applied for network management tasks, especially Quality-of-Service (QoS) management.

KW - Network Management

KW - Next Generation Networks (NGN)

KW - Bayesian Belief Networks (BBN)

KW - Machine Learning.

M3 - Conference contribution

SN - 978-1-902560-22-9

BT - Unknown Host Publication

ER -