A study on comparison between VBR and CBR video service in ATM environment

The authors compare constant bit rate (CBR) and variable bit rate (VBR) video services in an asynchronous transfer mode (ATM) environment. The simplified network model of CCITT SGXV is used to estimate the statistical multiplexing gain for VBR service. Current NNI (network node interface) capacities, 150 or 600 Mb/s, are determined to be too small to effectively support standard television signals that are VBR coded at average bit rates of 5-10 Mb/s. It is also shown that the leaky bucket technique is superior to the sliding window technique for an average bit rate usage parameter control from the service point of view. If sliding window is used, higher picture quality and shorter delay are not always guaranteed in VBR video services.<
>