Multi-objective Optimization of Bandwidth Broker with Evolutionary Algorithm

The Optical Transport Networks with dynamic transport routing capability enable rearrangement of the logical link capacities on demand in real time. In this paper an evolutionary algorithm based approach for multi-objective optimization of Bandwidth Broker (BB) is presented. BB is represented as evolutionary process in dynamically changed traffic demand environment. Multi-objective Evolutionary algorithm scheme for bandwidth allocation is described. The suitability of Evolutionary algorithm approach is investigated. The Pareto optimal solutions are ensured by the underlying genetic operators. There are carried out a few experiments and the test results illustrate the trade-off between objectives and ability of this approach to produce many good compromise solutions. Ill.3, bibl.8 (in English; summaries in Lithuanian, English and Russian).

[1]  Eckart Zitzler,et al.  Evolutionary algorithms for multiobjective optimization: methods and applications , 1999 .

[2]  Joachim Charzinski,et al.  Evaluation of Effective Bandwidth Schemes for Self-Similar Traffic , 2000 .

[3]  Oliver Heckmann,et al.  Robust bandwidth allocation strategies , 2002, IEEE 2002 Tenth IEEE International Workshop on Quality of Service (Cat. No.02EX564).

[4]  Sam Kwong,et al.  A Genetic Algorithm For Joint Optimization Of Spare Capacity And Delay In Self-healing Network , 2002, GECCO.

[5]  Passakon Prathombutr,et al.  Virtual topology reconfiguration in wavelength-routed optical networks , 2003 .

[6]  Rudolf Mathar,et al.  An off-line traffic engineering model for MPLS networks , 2002, 27th Annual IEEE Conference on Local Computer Networks, 2002. Proceedings. LCN 2002..

[7]  Rajeev Kumar,et al.  Topological design of communication networks using multiobjective genetic optimization , 2002, Proceedings of the 2002 Congress on Evolutionary Computation. CEC'02 (Cat. No.02TH8600).