Analysis of effective fuzzy-CAC solution for proactive traffic engineering

Abstract

Actual Connection Admission Control (CAC)implementation inside the RSVP-TE protocol in MPLS-TE networks is highly limited, since the applied threshold CAC cannot provide Quality of Service (QoS) aware decision making on MPLS-TE network nodes. This prevents an effective end-to-end QoS control in a fully dynamic, application driven Label Switched Path (LSP) setup scenario. Authors of this paper have developed effective and flexible Fuzzy logic driven CAC solution (Fuzzy-CAC) which is capable of sustaining the parameters of QoS within the accepted limits and provide the dynamic setup of an application-controlled LSP. In this paper we present basic results as well as more in details discuss several adaptation approaches that help Fuzzy-CAC to operate under rapidly changing traffic characteristics.

Admission Control Algorithm for MPLS-TE Networks

“…CAC increases the availability of RSVP-TE to achieve high performance, especially for realtime applications, which need high end-to-end QoS. The authors in [9] introduced Fuzzy-CAC to work with dynamic traffic in the fast changing environment. Edge Admission Control is developed in [10] that computing admission thresholds and takes into account both the best-effort traffic's performance and QoS traffic's constraints.…”