Synthesizing transient traffic of temporal visual signals for discrete event simulation

This paper presents the analysis and modeling of transient traffic source of temporal visual signals for discrete event simulation. Transient nature implies the inability of applying established traffic theory that assumes stochastic processes. The focus is on the transient traffic of foreground object due to its motions as well as camera panning and zoom. We show that such transient traffic can be well modeled as power functions. Implemented for discrete event simulation in DEMOS, the model with a number of parameters can synthesize transient traffic of temporal visual signal in piecewise manner. Our simulation results reveal that arbitrary form of transient traffic can be synthesized in discrete event simulation as a series of the power function model with different parameter values. An exemplary application of the model in the simulation of a complex queuing system is given within our current research in futuristic multi-camera tele-immersive collaboration system.