A constrained optimization algorithm in split-spectrum techniques

The ultrasonic non-destructive evaluation is intended to detect flaws or defects in materials, analyzing the echoes produced by its internal structure. Many algorithms based on the Split-Spectrum Technique (SST) (i.e. splitting the original scan and compounding the different channels in several ways), have been developed to provide an improvement of the Signal to Noise Ratio (SNR). One of these algorithms, in which compounding is performed by choosing the minimum squared signal at each range has been shown to provide considerable improvement in SNR. In the present work, we have designed an optimum order statistic filter (OSF), in order to provide the maximum available Signal to Noise Ratio Enhancement (SNRE). The OSF computed is found to be very close to the minimum filter, and so, we have established a new approach to the minimization algorithm.