Ad hoc Networking, Markov Random Fields, and Decision Making

Abstract

Our central proposition is that by articulating decision making in an ad hoc network using a Markov random field (MRF) maximum a posteriori (MAP) approach, a viable decision- making framework can be designed. This article explains why distributed decision making in ad hoc networks is so challenging. It describes the MRF-MAP decision-making framework and in doing so highlights the many insights gained from image processing. The framework is explored in the context of deciding between different ad hoc routing protocols.