A novel generative model for burst error characterization in Rayleigh fading channels

Generative models have significant applications in the design and performance evaluation of communication protocols as well as coding systems. In this paper, a novel deterministic process based generative model (DPBGM) is proposed for the characterization of burst errors in Rayleigh fading channels. A simple procedure has been developed for generating an error sequence with desired burst error statistics from a properly sampled and parameterized deterministic process followed by a threshold. The proposed generative model enables us to perfectly match any given error burst distribution (EBD) and error-free burst distribution (EFBD) of the underlying descriptive model. The gap distribution (GD) and the error-free run distribution (EFRD) obtained from the generated error sequence can also be fitted to those of the target error sequence with good accuracy.