Keywords: fuzzy logic controller, simulation, differential evolution, control system, ATM
Received May 12, 2005; revised September 5, 2005 & January 24, 2006; accepted March 1, 2006.
Communicated by H. Y. Mark Liao.
| Previous | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 |
¡@
France Cheong, Richard Lai*, Yong Soo Kim+, Dae Hoon Hwang+, Hae Suk Oh+ and Ji Hong Kim+
School of Business Information Technology
RMIT University
Victoria 3000, Australia
*Department of Computer Science and Computer Engineering
La Trobe University
Victoria, 3086, Australia
+College of Software
Kyungwon University
Songnam, Kyunggi-Do, 461-701, Korea
Available Bit Rate (ABR) streams are carried whenever bandwidth is available to
do so. When transporting MPEG streams over ATM networks, the link utilization cannot
be brought close to 100% no matter how efficient the connection control admission
scheme is, but can be improved by ABR traffic. Due to the unpredictable fluctuations
and burstiness of the MPEG streams, it is difficult to accurately control the ABR flow
rate. A traffic control mechanism based on fuzzy logic is therefore required to effectively
transport ABR traffic so that network operators can use most of the unused bandwidth.
In this paper, we present the design of an optimized Fuzzy Logic Controller (FLC) of an
ATM switch for the rate control of an ABR stream. Unlike the previous work where
synthetic data and trial and error were used for tuning an FLC, the merits of our method
are that we used simulations and real MPEG data to generate the input-output data for
the optimal design of the FLC, and that our technique enables the output of the FLC to
be symmetrical.
Received May 12, 2005; revised September 5, 2005 & January 24, 2006; accepted March 1, 2006.
Communicated by H. Y. Mark Liao.