Neural call admission control through virtual links estimates

A neural call admission control (NCAC) decides whether a new connection can be accepted, according to a neural network decision. Such a decision is based on traffic information (e.g. cell loss rate (CLR)) measured in real time, at the ATM node, which ensures continual training of an artificial neural network (ANN) during NCAC performance, allowing the CAC to adapt to possible changes in traffic conditions. This paper addresses the problem of providing accurate CLR information to perform optimal ANN training, which results in efficient NCAC performance. This paper proposes a novel estimation method based on measuring the CLR at virtual links with different and slower output rates. This information can be related to the real CLR, by means of an ANN, thus solving the accuracy and estimation-duration problems of real-link estimates. Prior information which allows the ANN to interpolate the real CLR is also required to establish the relationship between the virtual and the real CLRs. This information has been named zero loss bandwidth patterns