C2 Organization Analysis Based On Complex Networks Theory

As an important branch of complexity science, complex networks provide a new viewpoint for investigation of complex systems. In recent years, with discovery of some common properties of real-world networks, such as small-world characters and free-scale characters, research on complex networks has attracted considerable interest in the scientific community and becomes a new inter-discipline. Similarly, in the military field, using latest theory and achievements of complex networks on research of networked warfare of information age, has received more and more concern.Because of the complexity essence of warfare, the warfare systems can not be studied by traditional reductionism. Concerning with the well-know problem of evaluating combat system such as command and control system, this paper suggests that use network topology to abstract and describe combat systems, evaluate the effectiveness of combat systems and study methods to optimize combat network.Firstly, we introduce the relationships between complexity and complex networks, the general situation of complex networks study, as well as it's process in military field. We also point out the significance of using network analysis methods to study complexity of warfare.Secondly, we discuss several important complex networks models in which regular model,random model,small-world model and scale-free model are included. The properties of these models are analyzed from aspects of average path length,clustering coefficient and degree distribution. At last, we use MATLAB to realize a simplified BA network model and test it's scale-free character.Thirdly, the FINC method is introduced as a quantitative approach for modeling and analyzing networked military systems. We point out the FINC method is a linear analysis and make some reasonable improvement to it according to the nonlinear factors of battlefield environment and warfare systems. We use improved FINC method to optimize and evaluate effectiveness of static military organization models.Finally, Adopting the FINC method as theoretical basic to construct combat simulation scenario, we present four typical paradigms of topological structure of C~4ISR systems, in which centralized model,distributed model,centralized with information sharing model and distributed with information sharing model are included. We use multi-agent based simulation software MASON to evaluate the effectiveness of these models. By comparing and analyzing the experimental results, we derive some basic principles for constructing military C2 organizations.