Title: | A Teletraffic Perspective on Relay-Node Selection Strategy in VoP2P System |
Authors: | Jia, Wen-Kang Chen, Yaw-Chung 資訊工程學系 Department of Computer Science |
Keywords: | VoP2P;GoS;Teletraffic;Relay-node;Blocking Rate;relay-node Handoff;S(2)RS;MRS |
Issue Date: | 2009 |
Abstract: | This paper gives a theoretical approach to modeling and simulating the Grade of Service (GoS) of Voice over Peer-to-Peer Session Initiation Protocol system which is created based on various teletraffic parameters. VoP2P features a particular service characteristic due to its different GoS models originated from circuit switched telephone networks. Especially, most voice session in VoP2P is indirectly connected because peer-nodes usually locate behind NATs. This situation forces peer-node to select one or more relay-node for establishing sessions. However, the relay-node is a constrained resource because every relay-node can only support very limited relay services in VoP2P network. There are two strategies, the S(2)RS (Staged Single-Relay Strategy) and MRS (Multi-Relay Strategy), to solve the relay-node issue. We adopt three new parameters in VoP2P call-loss models, the first is callee-attendance rate, which is due to the characteristic of destination peer-nodes with a lower attendance probability; the second is call-blocking rate, which is a traditional definition just because the relay-node reaches the limitation of resource; the third is handoff-dropping rate (aka handoff-loss rate), which is caused by the relay-node being removed frequently. The simulation results provide useful reference in the study of teletraffic of VoP2P system. |
URI: | http://hdl.handle.net/11536/16758 http://dx.doi.org/10.1109/NAS.2009.27 |
ISBN: | 978-0-7695-3741-2 |
DOI: | 10.1109/NAS.2009.27 |
Journal: | NAS: 2009 IEEE INTERNATIONAL CONFERENCE ON NETWORKING, ARCHITECTURE, AND STORAGE |
Begin Page: | 140 |
End Page: | 147 |
Appears in Collections: | Conferences Paper |
Files in This Item:
If it is a zip file, please download the file and unzip it, then open index.html in a browser to view the full text content.