Browsing by Author "Wang, Wei-Chien"

Now showing 1 - 2 of 2

An Effective Fair Bandwidth Allocation in Resilient Packet Ring
(2010)
Tang, Wen-Shiang
;
Wang, Wei-Chien
;
Chang, Chung-Ju
;
Hsieh, Yao-Ting
;
電機工程學系
In this paper, we propose an effective fuzzy local fairRate generator (FLFG) for resilient packet ring (RPR) to achieve better utilization fairness and convergence time. The FLFG is composed of three components: adaptive fairRate calculator (AFC), fuzzy congestion detector (FCD), and fuzzy fairRate generator (FFG). AFC produces an estimated fairRate in the meantime FCD indicates the congestion degree of station. Finally, FFG adopts the two outputs from AFC and FCD to generate precise local fairRate. Simulation results show that FLFG performs a stable fairness algorithm and improves at least about by 200% in converge time of the fairRate over the aggressive mode (AM) and the distributed bandwidth allocation (DBA) algorithms.
FLAG: A Fuzzy Local FairRate Generator for Resilient Packet Ring
(2010-05-01)
Tang, Wen-Shiang
;
Chang, Chung-Ju
;
Tien, Po-Long
;
Wang, Wei-Chien
;
電機工程學系
A local fairRate generator using fuzzy logic and the moving average technique is proposed for the resilient packet ring (RPR). The fuzzy local fairRate generator (FLAG) is designed to achieve both low convergence time and high system throughput, besides fairness. It contains three functional blocks, an adaptive fairRate calculator (AFC) to properly preproduce a local fairRate by the moving average technique, a fuzzy congestion detector (FCD) to intelligently estimate the congestion degree of the station, and a fuzzy fairRate generator (FFG) to precisely generate the local fairRate. Simulation results show that only the FLAG can stabilize all flows in parking lot scenarios with different finite traffic demands, compared with the conventional aggressive mode (AM) and distributed bandwidth allocation (DBA) fairness algorithms. Also, it attains a convergence time lower than the AM fairness algorithm by at least 7 times and the DBA fairness algorithm by at least 2 times in parking lot scenarios with greedy traffic demands.