Abstract

In this paper, two distortion measures, based on the well known statistical Log Likelihood Ratio (LLR) criterion are presented. The first one is developed for non-stationary scalar processes, represented by a Time Varying Autoregressive (TV AR) model and the second for stationary vector processes, represented by a Multichannel Autoregressive (MC AR) model. The asymptotic distribution of the proposed measures is briefly discussed.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call