Multicast security: a taxonomy and some efficient constructions

TLDR

Multicast communication is increasingly used, yet existing security protocols provide only partial protection and authenticating sources is more complex and inefficient than in unicast. The paper aims to develop comprehensive security mechanisms for multicast, including a taxonomy of scenarios and efficient solutions for source authentication and key revocation. The authors propose a taxonomy of multicast scenarios and introduce a midpoint approach between MACs and digital signatures for source authentication, along with an improved key‑revocation scheme.

Abstract

Multicast communication is becoming the basis for a growing number of applications. It is therefore critical to provide sound security mechanisms for multicast communication. Yet, existing security protocols for multicast offer only partial solutions. We first present a taxonomy of multicast scenarios on the Internet and point out relevant security concerns. Next we address two major security problems of multicast communication: source authentication, and key revocation. Maintaining authenticity in multicast protocols is a much more complex problem than for unicast; in particular, known solutions are prohibitively inefficient in many cases. We present a solution that is reasonable for a range of scenarios. This approach can be regarded as a 'midpoint' between traditional message authentication codes and digital signatures. We also present an improved solution to the key revocation problem.