Covert channels-here to stay?

We discuss the difficulties of satisfying high-assurance system requirements without sacrificing system capabilities. To alleviate this problem, we show how trade-offs can be made to reduce the threat of covert channels. We also clarify certain concepts in the theory of covert channels. Traditionally, a covert channel's vulnerability was measured by the capacity. We show why a capacity analysis alone is not sufficient to evaluate the vulnerability and introduce a new metric referred to as the "small message criterion".<
>