A Hybrid Approach to Multicast Data Origin Authentication

Multicast is an efficient communicating scheme in group communications, and data origin authentication is one of the most important problems in multicasting. But at present, there is no ideal approach that is efficient for all sorts of multicast applications. For example, tree-hashing´s authentication probability can maintain 100%, but it has high communication overhead. Chain-hashing´s computation overhead and communication overhead are low, but it isn´t efficient when any data packet is lost or disordered in transit. Hence, this paper proposes a Hybrid Approach to Multicast data origin Authentication (HAMA), which is based on tree-hashing and chain-hashing. Approved by the performance evaluations and simulation results, HAMA can support non-repudiation, has less communication overhead than that of the traditional tree-hashing, and the authentication probability of HAMA is much higher than that of the traditional chain-hashing, since it can maintain 90% if the loss probability is less than 25%. Thus, HAMA is secure, reasonable, and efficient, especially for multicasting unreal-time data.