A novel group key transfer for big data security

Abstract

With the rapid development of group-oriented services over big data, it needs novel solutions to ensure big data security. Most existing group key transfer protocols depend on a mutually trusted key generation center (KGC) to transport the group key to all group members secretly. This approach requires a trusted sever to be set up, and it incurs communication overhead costs. In addition, existing protocols based on secret sharing scheme use traditional threshold schemes to compute a t-degree interpolating polynomial and uses the polynomial to encrypt and decrypt the secret group key. This approach increases the computational complexity. In this paper, we construct efficient and secure group key transfer based on secret sharing over big data. We first present a novel group key transfer protocol without an online KGC, which is based on DH key agreement and a perfect linear secret sharing scheme (LSSS). The confidentiality of the group key transfer phase of this protocol is information theoretically secure, which is ensured by the LSSS. Furthermore, this protocol can resist potential attacks and also significantly reduces the overhead of system implementation. Such a protocol is desirable for many group-oriented applications over big data.