Analysis of the Noise Immunity of Emission Prediction in the Dynamics of the Heartbeat Using Information about Short-Term and Long-Term Dependencies

Abstract

A comparative analysis of the effectiveness of prediction using PRT and RIA approaches, using, respectively, exclusively the linear component of long-term memory and, along with the linear, non-linear component, is given. The noise immunity of prediction is considered in both approaches in the presence of additive noise with a normal or uniform distribution.