In this paper, we propose the use of a neural-fuzzy scheme for a rate-based feedback controller in ATM (Asynchronous Transfer Mode) networks. Traditional methods perform congestion control by monitoring the queue length. When the queue length is greater than a predefined threshold, the source rate is decreased by a fixed rate. However, the determination of the threshold and the rate is a big problem associated with these methods. Also, some users cannot obtain the fair share at a switch or network element because of the constraints imposed by the limited amount of bandwidth available at other switches along its path. Our system can overcome these problems. Possible traffic congestion is detected by monitoring the actual traffic flow. An ART-like neural-fuzzy network is employed to predict the future cell loss which is treated as a parameter and the source rate is calculated based on a fairness algorithm that can evaluate the fair share for users. Simulation results have shown that our method is more effective than traditional methods.
[1]
Chung-Ju Chang,et al.
Design of a fuzzy traffic controller for ATM networks
,
1996,
TNET.
[2]
Jyh-Shing Roger Jang,et al.
ANFIS: adaptive-network-based fuzzy inference system
,
1993,
IEEE Trans. Syst. Man Cybern..
[3]
Chin-Teng Lin,et al.
An ART-based fuzzy adaptive learning control network
,
1997,
IEEE Trans. Fuzzy Syst..
[4]
Semyon M. Meerkov,et al.
Feedback control of congestion in packet switching networks: the case of a single congested node
,
1993,
TNET.
[5]
Christos Douligeris,et al.
Rate Regulation with Feedback Controller in ATM Networks - A Neural Network Approach
,
1997,
IEEE J. Sel. Areas Commun..
[6]
K. K. Ramakrishnan,et al.
Time scale analysis scalability issues for explicit rate allocation in ATM networks
,
1996,
TNET.
[7]
C. Douligeris,et al.
Neural network-based congestion control for single and multiple-sources in ATM networks
,
1996
.