Autonomic defense: Thwarting automated attacks via real‐time feedback control
D. A. Armstrong, Sam Carter, Gregory L. Frazier, Tiffany M. Frazier · Complexity · 2003
Abstract A critical threat to organizations, and the Internet itself, is a class of automated network attacks referred to as Internet worms. This article examines the use of mathematical models and optimization algorithms—specifically a partially‐observable Markov decision process (PO‐MDP) based feedback control system—as the basis for implementing an autonomic defense system (ADS) that can protect organizations against Internet worms. The PO‐MDP ADS introduced in this article is capable of detecting and responding to worms in real time. Furthermore, the PO‐MDP ADS can ameliorate the rate of incorrect control decisions that would normally occur in the presence of sensor false alarms. © 2004 Wiley Periodicals, Inc. Complexity 9: 41–48, 2004