Blind deconvolution of linear systems driven by non-stationary source based on almost-symmetrical time-varying ARMA model

Two blind convolution schemes based on an almost-symmetrical time-varying autoregressive moving average (ARMA) model are proposed. Such a model is constructed for analytical source signals. Blind deconvolution of two categories of sources (constant-envelope and time-varying envelope) are discussed. The uniqueness of the blind deconvolution is shown to be guaranteed.<>