A hybrid genetic algorithm/variable neighborhood search approach to maximizing residual bandwidth of links for route planning

Kandavanam, Gajaruban and Botvich, Dmitri and Balasubramaniam, Sasitharan and Jennings, Brendan (2010) A hybrid genetic algorithm/variable neighborhood search approach to maximizing residual bandwidth of links for route planning. In: Artificial Evolution - 9th International Conference Evolution Artificielle, EA 2009, Revised Selected Papers :. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) . UNSPECIFIED, FRA, pp. 49-60. ISBN 3642141552

Full text not available from this repository. (Request a copy)

Abstract

This paper proposes a novel approach to performing residual bandwidth optimization with QoS guarantees in multi-class networks. The approach combines the use of a new highly scalable hybrid GA-VNS algorithm (Genetic Algorithm with Variable Neighborhood Search) with the efficient and accurate estimation of QoS requirements using empirical effective bandwidth estimations. Given a QoS-aware demand matrix, experimental results indicate that the GA-VNS algorithm shows significantly higher success rate in terms of converging to optimum/near optimum solution in comparison to pure GA and another combination of GA and local search heuristic, and also exhibits better scalability and performance. Additional results also show that the proposed solution performs significantly better than OSPF in optimizing residual bandwidth in a medium to large sized network.

Item Type:	Book Section
Uncontrolled Keywords:	/dk/atira/pure/subjectarea/asjc/2600/2614
Departments or Groups:
Depositing User:	Admin SSL
Date Deposited:	19 Oct 2022 23:16
Last Modified:	08 Jun 2023 01:35
URI:	http://repository-testing.wit.ie/id/eprint/5031

Actions (login required)

View Item