Data Mining for Managing Intrinsic Quality of Service in MPLS

Abstract

LSP set up admission control policy is one of the notable problems that have to be solved to fulfill the requirements for effective resource allocation and network utilization for appropriate QoS level. In this paper, we verify a possibility of a new LSP setup admission algorithm, which uses optimization procedure based on multi-objective model with Pareto ranking and Genetic Algorithm. Decision rules are generated with Data Mining approach by performing classification operation to the selected data. This algorithm functions in two phases – classification and operating, which are accomplished consecutive. Algorithm is described and depicted. Experimental data are depicted and future research subjects are pointed. Ill. 4, bibl. 15 (in English; summaries in English, Russian and Lithuanian).