Distributed belief propagation for situation assessment in net-centric environment

A distributed net-centric environment consist of a large variety of data fusion nodes, where each node represents a sensor, software program, machine, human operator, warfighter, or a combat unit. Fusion nodes can be conceptualized as intelligent autonomous agents that communicate, coordinate, and cooperate with each other in order to improve their local situational awareness (SA), and to assess the situation of the operational environment as a whole. In this paper, we describe how we model this net-centric SA problem using a distributed belief propagation paradigm. A local fusion node maintains the joint state of the set of variables modeling a local SA task at hand using Bayesian network (BN) fragments. Local fusion nodes communicate their beliefs and coordinate with each other to update their local estimates of the situation and contribute to the global SA of the environment. We have implemented the propagation paradigm to determine threat out of terrorist dirty bombs with agents searching unstructured intelligence reports for evidence and assessing local situations via BN fragments. The paradigm provides an important foundation of our company's cutting-edge predictive analytics platforms offering to solve enterprise distributed big data search and analytics problems.