Multicast Data Security in PIM-SM

Multicast has been a topic of research for over two decades , emerging from drawbacks in unicast and broadcast networks. In multicast, users can be part of specific groups and communicate group information amongst themselves besides being part of a larger network. There has been successful research in its addressing system, architecture, routing protocols, but security continues to remain an area which needs more attention. Security in multicast is more complicated as compared to unicast, as latter deals with only a pair of senders and receivers against multicast, dealing with 1:1 or 1:N or N:N senders and receivers. Maintaining security in such dynamic networks, is one of the most important challenges amongst research groups. Today, many key management schemes or security architectures have been proposed and standardized, but have not been applied to all existing multicast protocols. With ever increasing demand of multicast in commercial and military applications, it is important to have a more reliable, trustworthy and secure group to communicate. This thesis, investigates a solution for data message security in one of the widely used multicast protocols, PIM-SM (Protocol Independent Multicast - Sparse Mode). PIM-SM is more vulnerable in terms of security as it supports members scattered throughout a network or spanning over a wide area. Currently, PIM-SM specification suggests security for its control messages but does not support any security for data messages exchanged within a group. We propose using a standard key management protocol GSAKMP (Group Secure Association Key Management Protocol) for a more efficient and secured multicast protocol. The thesis initially discusses data security concerns in PIM-SM. It provides a concept of operation explaining integration of PIM-SM and GSAKMP to develop data security between senders and multicast group members in PIM-SM. The solution maintains individual architectures and explains how the two can be used as a combined architecture