Maintaining authenticated communication in the presence of break-ins

We study the problem of maintaining authenticated communication over untrusted communication channels, in a scenario where the communicating parties may be occasionally and repeatedly broken into for limited periods of time. Once a party is broken into, its cryptographic keys are exposed and perhaps modified. We describe a mechanism that allows a party whose security has been compromised to regain its ability to communicate in an authenticated way. The contribution of this paper is twofold. First we present a mathematical model for analyzing this scenario, and exhibit various properties and parameters of this model. Next we describe a practically-appealing protocol which enables parties to maintain authenticated communication in the presence of such a powerful adversary. For this protocol we use a variation of the proactive distributed signature schemes which were recently described by Herzberg et al. Although these schemes are designed for a model where authenticated communication and broadcast primitives are available, we show how they can be modified to work in our model, where no such primitives are available a-priori. We also present a new proactive distributed signature scheme with improved round and communication complexities.