A. Parvaresh, M.I. Kalantari, and A. Movaghar (Iran)
Connection Admission Control, Case Based Reasoning, Fuzzy Logic, Quality of Service, Artificial Intelligence.
In recent high speed networks such as ATM, dealing with different kinds of traffics have become a promising goal.Due to the dynamic nature and burstiness of aggregation of these various traffics, which is shown to be self similar, network congestion would occur more probably. Thus congestion control mechanisms (e.g. connection admission control) should be applied to preserve the quality of service of traffics passing through the network. In this paper we present a connection admission control (CAC) algorithm for self similar traffics based on case based reasoning (CBR) which is almost a new concept in applied artificial intelligence. CBR is used to correct the estimation of the number of admitted sources to a link to achieve more link utilization than conventional CAC schemes which is known to underestimate the number of admitted sources. At last we introduce fuzzy logic concept to the proposed system to improve the CBR CAC performance.
Important Links:
Go Back
