Enhanced Green Bandwidth Allocation by Improving Intra-ONU Scheduling in Ethernet Passive Optical Networks

  • 林 慈洋

Student thesis: Master's Thesis


In order to ensure the energy efficiency of each optical network unit (ONU) in an access network and guarantee quality of service (QoS) requirements of differentiated services such as Voice over IP (VoIP) Video on Demand (VoD) real-time near real-time transactions and data traffic the Ethernet Passive Optical Network (EPON) is considered the solution Many studies have proposed sleep mechanisms that concern a single class delay constraint While there exist multiple classes of service on a network transmission saving energy under different classes considered causes the problem that with the stringent delay constraint classes considered causes the less energy saving and stringent class will violate their delay constraint while considered with the relaxed delay constraint classes The authors of Green Bandwidth Allocation (GBA) proposed a new analytical model to determine the optimal sleep time of ONUs in every cycle without violating the maximum class delay constraint Although the GBA mechanism guarantees that the different class delay constraints do not violated the maximum class delay constraint packets with a more relaxed delay constraint will be treated as the most stringent delay constraint and may be sent early This means that the ONU will waste time in active mode to send packets which did not need to be sent at the current time Accordingly we proposed a novel model in intra-ONU scheduling to control the packets under different delay constraints thereby enhancing GBA performance Moreover packets in the same classes may have different packet delay constraints We improve the classifier on Enhanced GBA to classify packets according to their packet delay constraint Further we use the new ONU architecture in which the ONU can switch the architecture according to the sleep time length thus improving energy saving in the Enhanced GBA based on Enhanced Classifier The simulation results show that the Enhanced GBA and Enhanced GBA based on Enhanced Classifier ensure that different class packets do not violate their delay constraint and consume less energy than GBA in different cases The Enhanced GBA based on Enhanced Classifier can be used in traffic where the delay constraint is not according to classes Instead packets in same class also have different packet delay constraints
Date of Award2015 Aug 21
Original languageEnglish
SupervisorChuan-Ching Sue (Supervisor)

Cite this