PARALLEL ALGORITHMS FOR DIGITAL PROCESSING OF GEOPHYSICAL SIGNALS

Abstract

This article presents the development of parallel computational approaches for spectral analysis based on two-dimensional fast Haar transform algorithms in the digital processing of geophysical signals. The study analyzes the processing of signals obtained through magnetometry, gravimetric, and electrical prospecting methods using spline functions and the Haar transform. A parallel computing technique based on bipolynomial basis functions and fast transform algorithms was developed and implemented on ADSP-BF533 and BF-561 processors, with a comparison of their computational speeds. The results demonstrate that dual-core processors achieve significantly higher performance compared to single-core processors.