This paper presents GREP, a novel group rekeying scheme that leverages the history of join events in order to achieve efficiency and high scalability. GREP rekeys the group with only two broadcast messages, hence displaying an overhead which is small, constant and independent of the group size. Also, GREP efficiently recovers the group from collusion attack with no recourse to total member reinitialization. Even in the very unlikely worst case, collusion recovery displays a smooth impact on performance that gradually increases with the attack severity. We implemented GREP for the Contiki OS and tested it on different resource-constrained platforms. Our analytical and experimental evaluation confirms that GREP is efficient, highly scalable and deployable also on constrained nodes. The paper extends a previous version of this work, especially through additional security analysis, treatise of probabilities for worst case collusion, and experimental evaluation of performance.
Group rekeying based on member join history
Gianluca Dini
Secondo
;
2019-01-01
Abstract
This paper presents GREP, a novel group rekeying scheme that leverages the history of join events in order to achieve efficiency and high scalability. GREP rekeys the group with only two broadcast messages, hence displaying an overhead which is small, constant and independent of the group size. Also, GREP efficiently recovers the group from collusion attack with no recourse to total member reinitialization. Even in the very unlikely worst case, collusion recovery displays a smooth impact on performance that gradually increases with the attack severity. We implemented GREP for the Contiki OS and tested it on different resource-constrained platforms. Our analytical and experimental evaluation confirms that GREP is efficient, highly scalable and deployable also on constrained nodes. The paper extends a previous version of this work, especially through additional security analysis, treatise of probabilities for worst case collusion, and experimental evaluation of performance.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
