Estimation for Nonstationary Pressure Signals

This paper presents a new Kalman filter-based Recently, the parametric spectrum of pressure signals, power spectral density estimation (PSD) algorithm for including arterial, intracranial, cardiac pressure, etc. has nonstationary pressure signals. The pressure signal is assumed been studied by Aboy et al. [4, 5]. They constructed a novel to be an autoregressive (AR) process, and a stochastically statistical model to describe the frequency variability of perturbed difference equation constraint model is used to pressure signals. In this synthetic pressure signal model, the describe the dynamics of the AR coefficients. The proposed effects of respiration were incorporated on arterial blood or Kalman filter frame uses variable number of measurements to intracranial pressure. Subsequently, a Kalman filter-based estimate the time-varying AR coefficients and yield the PSD PSD estimate algorithm, which is similar to [I]-[3], was estimation with better time-frequency resolution. Simulation developed by Aboy et aL for the synthetic pressure signals results show that the proposed algorithm achicvcs a better tim-freqults ncywthat rtsoutionosthanlconnti algorithms ftr [5]. Their spectral density estimate had an unsatisfactory time-frequency resolution than conventional algorithms for rslto nfeunydmi,bcuetecnetoa nonstationary pressure signals.resolution in frequency domain, because the conventional nonstationary pressure signals. Kalman filter only employed one measurement to update the AR coefficients so that the AR coefficient estimates had a