Conceptual Framework for Representing Human Behavior Characteristics in a System of Systems Agent-based Survivability Simulation-Intelligent Networks.

摘要

This research was conducted as a pilot study to model and simulate a mobile ad hoc network (MANET) as a system of systems (SoS) using selected human behavioral traits. A MANET as used in the research is a physics-based as well as a socio-technical system. As a sociotechnical system (or human-machine system), multiple humans and multiple radios interact through some digital information connectivity. Thus, it is important to consider human behaviors in modeling and simulation of MANET systems in order to obtain useful performance metrics. It is observed that behavioral models are useful as MANETs seek to discover their neighbors through interconnection which are dependent on many social-behavioral factors such as trust and cooperation. In this project, a MANET is represented and modeled as a symmetric nxn fuzzy matrix. An algorithm for calculating the evolution behavior of MANET nodes using fuzzy product operations on the matrix is developed. It is shown that after a sufficiently large number of iterations , the fuzzy matrix converges to a zero-one matrix. The values of 0 and 1 simply means that two edges of the network with 1 have a crisp connectivity (and hence good communication), and those with 0 means no connectivity at all.