Generation of Antibunched Light by an Inhibited Poisson Stream

A model of a conditioned Poisson stream of events is proposed in which the intensity is a Markov chain assuming the values f, g, 0 with 0 < f < g. The conditioned Poisson process is specified by assuming that an emission results in the intensity making a transition to the next lower value. In the absence of emissions, the intensity which is a process by itself has a propensity to move up in its state space, the residence time in any state being exponentially distributed with parameter n. It is shown that such a model of emissions possesses antibunching and sub-Poisson properties.