In Asynchronous Transfer Mode (ATM) networks, congestion can be caused by u
npredictable statistical fluctuations of traffic flows and fault conditions
within the network. If congestion: happens, then the network performance f
or the already established connection will decrease. ATM networks use the p
reventive congestion control mechanisms such as Usage Parameter Control (UP
C) and Connection Admission Control (CAC) to avoid congested conditions. Kn
owing that many sources in ATM networks have variable traffic stream with d
ifferent QoS characteristics, traffic management functions become necessary
to control the traffic flows within the network. By using the signaling pr
ocedures at the call setup phase, the network and source reach an agreement
for some traffic characteristic parameters. If the source violates the tra
ffic parameters, then the probability of congestion increases. So the netwo
rk must control the source traffic streams and detect well the violating ce
lls. Therefore, fast detection of airy violating: source is one of the most
important characteristics of a good traffic policer. In this paper we prop
ose a fuzzy traffic policer with high ability in detection of violating sou
rces. Our proposed fuzzy controller has two inputs, estimated passed mean c
ell rate and the current state of the counter. If the output of fuzzy contr
oller is 1, then the input cell is detected as violating cell, otherwise it
is a non-violating cell. Simulation results obtained from two traffic sour
ces, show that the proposed traffic policer has better selectivity than the
conventional leaky bucket. It is observed that our proposed traffic police
r has better ability for mean cell rate control, peak cell rate control and
burst duration control. Furthermore: it is observed that the proposed traf
fic policer outperforms the conventional leaky bucket specially when the dy
namic behavior is considered.