In this paper a class of model suitable for application to collision-sequence interference is studied. In these models it is assumed that the intervals between collisions are constant rather than exponentially distributed, as would be true if the collision times formed a Poisson process. The model may be two dimensional or three dimensional. Velocities are assumed to be completely randomized in each collision. The distribution of velocities is assumed to be Gaussian, though use is not always made of that fact. As applied to vector collisional interference the models allow the evaluation of the effects of windowing, which is of importance for $\mathcal{N}$-body simulation of more physically accurate models. They also lead to estimation of the effects of infilling of the interference dip following from deviations of the induced dipole moment from the intermolecular force. As applied to scalar collisional interference the models show the existence of a hitherto unknown (albeit weak) correlation between immediately successive collisions. An extension to the models, in which the magnitude of the induced dipole moment is equal to an arbitrary power or sum of powers of the intermolecular force, allows estimates of the infilling of the interference dip by the disproportionality of the induced dipole moment and force. One particular such model leads to the most realistic estimate for the infilling yet obtained.

• Received 3 March 2008

DOI:https://doi.org/10.1103/PhysRevA.77.062702