Fuzzy Flow Control and Its Implementation for ATM ABR Traffic

• Main Authors: Hsin-Tzung Wu, 吳信宗
• Other Authors: Yukon Chang
• Format: Others
• Language: en_US
• Published: 2000
• Online Access: http://ndltd.ncl.edu.tw/handle/55123840983805520267

碩士 === 義守大學 === 資訊工程學系 === 88 === A congestion control technique, named Ping-Pong flow control (PPFC), for ATM ABR traffic is presented in this article. It is a rate-based flow control scheme, where rate regulation is achieved by adjusting transmission rate in the source end station directly. The proposed algorithm uses a hybrid feedback strategy, called bipolar feedback, that employs mutually exclusive positive and negative feedback to control transmission rate under different switch states. These states are detected by fuzzy determination technique so that the issue of threshold selection can be avoided. We also introduce a new technique, called state early detection, that enables PPFC to control traffic flows more precisely and accurately at critical moments. We also offer an implementation of PPFC on both ATM switches and end stations. Simulation results show that the proposed algorithm has higher throughput and lower cell loss ratio when compared to the well-known BECN. These results also clearly demonstrate that our algorithm has excellent defense against feedback delay or even feedback loss.