According to the ITU-T Recommendation 1.371 [3], there are
two categories of control mechanism for ATM networks: pre
ventive and reactive controls. Preventive control is open-loop,
meaning the control algorithm makes decision based on its local
node, such as link capacity and buffer availability in the system.
Reactive control is close-loop in which network situations are
feedback from the network to the sources.
CAC is an example of preventive control. Its main role is to
determine whether a new connection request should be accepted
or rejected. The decision is based on the availability of resources
(bandwidth and buffer) and whether the OoS can be guaranteed
for a new connection without degrading the promiserl OoS of ex-.
isting connections. Conventional CAC relies on analytica[ mod
els in which parameters are based on a priori or user-specified
traffic descriptors [4] [5] [6]. However an analytical model may
not be able to capture the statistical nature of traffie sources.
This is because VBR traffic characteristics such as Sustainable
Cell Rate (SCR), Intrinsic Burst Tolerance (IBT) and Cell Delay
Variation (COY) are difficult for users to specify precisely. Im
proper specification of traffic characteristics could lead to either
degraded OoS or link under-utilization.