Application of packet assembly technology to digital video and VoIP

  • Authors:
  • Toshikatsu Kanda;Kazunori Shimamura

  • Affiliations:
  • Japan Telecom, Chuo-ku, Tokyo, Japan;Kochi University of Technology, Tosayamada Kochi, Japan

  • Venue:
  • Proceedings of the 12th annual ACM international conference on Multimedia
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

The Internet is composed of many kinds of networks and the networks are composed of network nodes such as routers. Routers use processor power for forwarding each packet with any size. At that time, node processor would be a bottleneck in respect to the high throughput if there would be too many packets to forward. Then, authors propose the packet assembly method. This aims to decrease the number of packets for the reduction of processor load, based on the fact that there are many packets much smaller than maximum transferable unit in backbone network. For the examination of the packet assembly, authors conducted two experiments. One is the experiment that conducts the packet assembly method for the traffic of digital video, and it provides the comparison of the image of digital video forwarded via routers without packet assembly with the one with packet assembly, and transition of edge router load and core router load. The other is the experiment that conducts the packet assembly method for the traffic of VoIP, and investigated about the influence on PSQM score, latency, and jitter.