Estimating efficiency of acoustic emission signal processing methods in implementation of polynomial digital filters

Abstract

The article presents a method for processing an acoustic emission (AE) signal to extract information and noise components from a noisy record. The method is based on digital filtering by polynomial models. To compensate for the introduced distortions by filters, a scheme for bidirectional processing of the AE signal is proposed. The operability of the filtering method for a noisy record is analyzed, and processing is evaluated on the basis of quantitative indicators. The evaluation results showed that the presented filtering method provides stability to the influencing noise and high accuracy of processing the AE signal recording when compared with the closest analogues.