Performance analysis of multicast algorithms for mobile satellite communication networks1The views expressed in this article are those of the authors and do not reflect the official policy of the United States Air Force, Department of Defense or the U.S. Government.1

  • Authors:
  • Ryan W Thomas;Richard A Raines;Rusty O Baldwin;Michael A Temple

  • Affiliations:
  • Department of Electrical and Computer Engineering, Air Force Institute of Technology, 2950 P St, Bldg 640, Wright-Patterson AFB, OH 45433, USA;Department of Electrical and Computer Engineering, Air Force Institute of Technology, 2950 P St, Bldg 640, Wright-Patterson AFB, OH 45433, USA;Department of Electrical and Computer Engineering, Air Force Institute of Technology, 2950 P St, Bldg 640, Wright-Patterson AFB, OH 45433, USA;Department of Electrical and Computer Engineering, Air Force Institute of Technology, 2950 P St, Bldg 640, Wright-Patterson AFB, OH 45433, USA

  • Venue:
  • Computer Communications
  • Year:
  • 2002

Quantified Score

Hi-index 0.24

Visualization

Abstract

With the rise of mobile computing and an increasing need for ubiquitous high-speed data connections, Internet-in-the-sky solutions are becoming increasingly viable. To reduce the network overhead of one-to-many transmissions, the multicast protocol has been devised. The implementation of multicast in these low earth orbit (LEO) constellations is a critical component to achieving an omnipresent network environment. This paper examines the system performance associated with two terrestrial-based multicast mobility solutions, distance vector multicast routing protocol (DVMRP) with mobile IP and on demand multicast routing protocol (ODMRP). These protocols are implemented and simulated in a satellite LEO constellation. Results from the simulation trials show the ODMRP protocol provided greater than 99% reliability in packet deliverability, at the cost of more than 8 bits of overhead for every 1 bit of data for multicast groups with multiple sources. In contrast, DVMRP proved robust and scalable, with data-to-overhead ratios increasing logarithmically with membership levels. DVMRP also had less than 70ms of average end-to-end delay, providing stable transmissions at high loading and membership levels.