We present algorithms for connections admission control (CAC) in an ATM net
work based upon the use of on-line measurement and estimation. (The algorit
hms described in this paper are the subject of a UK patent application.) Ou
r approach uses the theory of large deviations, which is a valuable tool fo
r reasoning about rare events in stochastic systems. We have shown previous
ly that the large deviations rate function, or entropy, of ATM traffic can
be estimated from activity measurements. This can be used to determine the
effective bandwidth of the traffic. In this paper we present practical meas
urement-based CAC algorithms which are computationally efficient, and can b
e used with traffic whose statistical characteristics are unknown. (C) 1998
Elsevier Science B.V. All rights reserved.