ChinaCom2008-Optical Communications and Networking Symposium

Research Article

QoS-Promoted Dynamic Bandwidth Allocation for Ethernet Passive Optical Networks

  • @INPROCEEDINGS{10.1109/CHINACOM.2008.4685014,
        author={Wen-Shiang Tang and Chin-Ya Huang and Chung-Ju Chang and Fang-Ching Ren},
        title={QoS-Promoted Dynamic Bandwidth Allocation for Ethernet Passive Optical Networks},
        proceedings={ChinaCom2008-Optical Communications and Networking Symposium},
        publisher={IEEE},
        proceedings_a={CHINACOM2008-OCN},
        year={2008},
        month={11},
        keywords={quality-of-service (QoS) dynamic bandwidth allocation Ethernet passive optical networks (EPONs) optical line terminal (OLT) and optical network unit (ONU).},
        doi={10.1109/CHINACOM.2008.4685014}
    }
    
  • Wen-Shiang Tang
    Chin-Ya Huang
    Chung-Ju Chang
    Fang-Ching Ren
    Year: 2008
    QoS-Promoted Dynamic Bandwidth Allocation for Ethernet Passive Optical Networks
    CHINACOM2008-OCN
    IEEE
    DOI: 10.1109/CHINACOM.2008.4685014
Wen-Shiang Tang1, Chin-Ya Huang1, Chung-Ju Chang1,*, Fang-Ching Ren2
  • 1: National Chiao Tung University, Hsinchu, Taiwan.
  • 2: Computer and Communication Research Labs, Industrial Technology Research Institute, Hsinchu, Taiwan.
*Contact email: cjchang@mail.nctu.edu.tw

Abstract

In this paper, we propose a QoS-promoted dynamic bandwidth allocation (Q-DBA) method to support quality-ofservice (QoS) requirement services for Ethernet passive optical networks (EPON). The Q-DBA method classifies the three types of services: voice, video, and data into six priorities. It can not only satisfy QoS requirements of real-time service and but also improve the QoS of non-real-time packets. Simulation results show that the Q-DBA can fulfill the video dropping probability requirement while the conventional dynamic bandwidth allocation with multiple services (DBAM) fails. Also, the Q-DBA can achieve the fairness index of average data delay close to 1 but DBAM’s fairness index of average data delay varies greatly.