Generation Mechanism for Cyclostationary and Self-Similar Processes

This paper proposes a generation mechanism for cyclostationary and self-similar processes. The proposed model extracts the information from the immediate coarser scale and adds the innovations to it to obtain the finer scale representation of the stochastic process. Basic block of the proposed model is the subband coder. By cascading the blocks of subband coder and passing white noise as one of input in addition to coarser information at each stage, cyclostationary process is generated. For generation of self-similar processes, white noise input for each stage is given in particular fashion. The mapping from finer scale to the immediate coarser scale is obtained using proposed blurring model. We have also given scheme for estimation of parameters of the proposed generation mechanism. The parameters are estimated for a given statistics case as well as the given data case. The proposed model can be used in variety of applications such as speech and image processing, biomedical signal processing, seismic data processing etc