engleski

Performance Analysis of the LPA-RICI Denoising Method

In this paper we have analyzed the performance of the recently introduced LPA-RICI signal denoising method. The RICI method is an extension of the classical denoising method based on the intersection of confidence intervals (ICI) rule, whereby an additional criterion, the ratio of intersection of confidence intervals, has been included. The performance of the RICI method is studied on a several test signals for an extensive range of noise distributions and SNR values. It is shown that for all types of noise and all considered SNR values, the RICI method when compared to the ICI method significantly reduces the estimation mean squared error.

Signal denoising; signal restoration; adaptive filtering; adaptive varying window selection; one-dimensional signal processing; relative intersection of confidence intervals (RICI) rule; intersection of confidence intervals (ICI) rule; local polynomial approximation (LPA).

nije evidentirano