A fitting report position scheme for the gated IPACT dynamic bandwidth algorithm in EPONs

Chuan Ching Sue, Hsaing Wen Cheng

研究成果: Article同行評審

22 引文 斯高帕斯(Scopus)

摘要

In EPONs using the gated Interleaved Polling with Adaptive Cycle Time (IPACT) scheme, the position of the report message within the granted transmission window has a direct effect on the average packet delay within the network. In this paper, this delay is minimized by using a fitting report position (FRP) scheme to adaptively adjust the position of the report message within the transmission window in accordance with the current network load. In the proposed approach, the optimal position of the report message is determined analytically for various system loads. The optical line terminal (OLT) then uses a heuristic algorithm to estimate the load of the optical network units (ONUs) in accordance with their report messages and determines the report message position that minimizes the average packet delay within the network. Finally, the OLT informs the ONUs of the optimal report position through an optional field in the gate message. The performance of the proposed FRP scheme is evaluated for three different network models, namely Poisson traffic with a uniform ONU load, Poisson traffic with a nonuniform ONU load, and self-similar traffic, respectively. The simulation results show that the FRP scheme achieves a lower average packet delay than fixed-report-position schemes such as fixed-report-front (FRF) or fixed-report-end (FRE) for both Poisson and self-similar traffic. The performance improvement is particularly apparent in networks with a nonuniform ONU load distribution.

原文English
文章編號5299244
頁(從 - 到)624-637
頁數14
期刊IEEE/ACM Transactions on Networking
18
發行號2
DOIs
出版狀態Published - 2010 四月 1

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Science Applications
  • Computer Networks and Communications
  • Electrical and Electronic Engineering

指紋 深入研究「A fitting report position scheme for the gated IPACT dynamic bandwidth algorithm in EPONs」主題。共同形成了獨特的指紋。

引用此