Neuro-fuzzy modeling and prediction of VBR MPEG video sources

The work presented in this paper intends to apply neuro-fuzzy methods for t he modeling and prediction on traffic intensity of digital video sources wh ich are coded with hybrid Motion Compensation/Differential Pulse Code Modul ation/Discrete Cosine Transform (MC:DPCM:DCT) algorithm. Although current c oding standards recommend constant bit rate (CBR) output by means of a smoo thing buffer, the hybrid algorithm inherently produces variable bit rate (V BR) output. This paper describes the novel application of a fuzzy predictor for the purposes of modeling and prediction on video sources. The computat ion requirement of the fuzzy predictor and its neural network implementatio n are also discussed. The proposed fuzzy prediction method and its neural n etwork version can be applied to the development of connection admission co ntrol, usage parameter control and congestion control algorithms in ATM net works. (C) 1999 Academic Press.