Signal conditioning for the recursive least-squares filter in ultrasonic testing of materials

This work aims to enhance the filtering performance of the recursive least-squares adaptive algorithm in the ultrasonic testing of materials by processing the primary input of the filter to improve the correlation properties of the signal and noise and enhance the adaptation and filtering process. The preprocessing is carried out using the SURE and universal threshold methods, which were originally used in wavelet denoising. The standard and proposed realisations of the recursive least-squares filter were applied to process simulated ultrasonic grain noise signals and the echo signals of experimental defects. Preprocessing using the SURE method did not offer good results in the simulation study. The results of the simulation and experimental studies showed that the proposed configuration of the recursive least-squares filter using the universal method performs better than the standard recursive least-squares filter.