A Comparative Analysis of Proposed Mobility Support Schemes for IP Multicast

Given the expeditionary nature of current and future Air Force operations, the Air Force will continue to rely on ad hoc mobile networks to accomplish its operational objectives. Mobile multicast technology will provide two major benefits. First, it will allow for the efficient the use of "push" technology for dissemination of mission critical information, and second, it will provide for improved coordination and control of operations involving entities dispersed and moving throughout the battle space. Allowing mobile hosts to connect to different links on an internet work while keeping the same IP address is the challenge of IP mobility. In RFC 2002, "IP Mobility Support", the Internet Engineering Task Force (IETF) defines the protocols and mechanism to provide IP mobility. Additionally, the IETF defines two mechanisms that allow mobile nodes to transmit multicast packets, and two mechanisms that allow mobile nodes to receive multicast packets. The transmission mechanisms are direct transmission, and home tunneling. The reception mechanisms are bi-directional tunneling and remote subscription. As proposed, the current IETF direct transmission mechanism requires the use of costly co-located care-of addresses, and does not provide for identification of senders' home IP addresses. This thesis proposes a novel modification to the IETF direct transmission mechanism. This modification, known as minimal multicast encapsulation, uses a modified form of minimal IP encapsulation to allow direct transmission while utilizing less costly foreign agent care-of addresses. Minimal multicast encapsulation also provides positive identification of sender IP addresses. In addition to demonstrating the viability of minimal multicast encapsulation, this research examines the performance of the four possible combinations of minimal multicast encapsulation or home tunneling with bi-directional tunneling or remote subscription. Comparisons are made in terms of path length, packet


loss,