As it gets more and more widely used on internet, multicast draws much attention for its security. Multicast distributed part of the system's load to internet, making a group of machines work for a single target. Meanwhile it brings security problems : how to protect a "logic" group , which contains lots of machines not in a single area or internet section ,as we do in unicast times? Finally these problems come down to the multicast cryto-key management protocol. In this paper I put forward a model that can be easily applied on existing internet system .This paper is organized in this way: section one details the problems multicast security meets; section two gives my model; section three makes a simple evaluation and comment on this model.
|