چکیده (انگلیسی):

Rapid growth of Internet spawns many group oriented multicast applications like Internet pay TV, news
dissemination and stock quote system. The fortes of these applications are the support of dynamic, scalable
group membership and group members are geographically divergent. As members of the group move in and
out, an imperative cryptographic rekeying model should be applied to preserve the confidentiality of the
group. A symmetric key called as session key is employed to defend the group communication data during
transit. Forward and backward secrecy is attained by updating the session key for every change in group
membership. Depends on the application immediate rekeying or batch rekeying can be used employed. The
problem with the batch rekeying algorithm is to determine the pertinent batch size and the optimal time for
rekeying process. The main aim is to propose a mathematical model based on queuing theory principles by
considering the request for rekeying as Poisson process, rekeying service as an exponential distribution. The
performance of the proposed model is analyzed using Java based simulator. By varying the arrival rate and
rekeying service rate the optimal batch size can be attained. The optimal rekeying interval improves the
performance of the group when the group membership grows dynamically. Reduces the long waiting time
of the rekeying requests and find the best batch size for the rekeying. Proposed mathematical model
analyses the various control parameters for batch rekeying and locates the best values for the batch size and
interval time using the M/M/1/K model queues.