Peak rate enforcement in ATM networks
IEEE INFOCOM '92 Proceedings of the eleventh annual joint conference of the IEEE computer and communications societies on One world through communications (Vol. 2)
Fast, approximate synthesis of fractional Gaussian noise for generating self-similar network traffic
ACM SIGCOMM Computer Communication Review
Asynchronous Transfer Mode Networks: Performance Issues
Asynchronous Transfer Mode Networks: Performance Issues
Genetic Algorithms in Search, Optimization and Machine Learning
Genetic Algorithms in Search, Optimization and Machine Learning
A VLSI fuzzy expert system for real-time traffic control in ATM networks
IEEE Transactions on Fuzzy Systems
IEEE Transactions on Fuzzy Systems
| Hi-index | 0.00 |
In ATM networks, usage parameter control is required in order to ensure that each source conforms to its negotiated parameters. This function is difficult to implement because of conflicting requirements such as selectivity and responsiveness. Traditional policing methods, such as leaky bucket and window-based mechanisms, have proved to be inefficient in coping with the conflicting requirements of ideal policing. A policing mechanism based on fuzzy logic has the capacity to combine a high degree of responsiveness with selectivity close to that of an ideal policer. However, the design of a fuzzy policer requires expert knowledge in the field.This project examines the applicability of genetic algorithms in the simultaneous design of membership functions and rule sets of a fuzzy traffic controller that performs policing in ATM networks. The proposed method leads to the development of rule sets and high performance membership functions for the fuzzy policer minimizing the need for human input in the design loop. The GA designs the fuzzy policer for various traffic conditions such as Poisson On-Off, TCP/IP and Self-Similar traffic. The performance of the GA-designed fuzzy policer is tested under real-time conditions using the NIST ATM Simulator. The reported simulation results show that the performance of the GA-designed fuzzy policer is much better than that of conventional policing mechanisms.