Statistical validation of a communications network simulation

Battlefield communications networks must deliver critical information when and where it is needed despite a rapidly changing and often hostile environment. Reliance upon computer simulations for system development and evaluation is often necessary since most communications systems are too complex to model analytically. Assurance that the simulation model faithfully emulates the process under study is essential in order to establish credibility and support the value of analyses and decisions that may follow. This paper describes a statistical procedure that provides an impartial assessment of agreement between simulated predictions and empirical observations for a communications network. The method is easy to understand, simple to implement, and flexible enough to hold the promise of more general and widespread application.