Some high efficiency digital signal processing techniques for high speed signal processing systems

Abstract

Some high efficiency digital signal processing techniques, based on digital filtering are presented for real-time general-purpose, high-speed signal processing systems. At first high efficiency one-dimensional and two-dimensional digital filters are described, both of non-recursive FIR type (using Cappellini window) and of recursive IIR type (using transformations). Further a special technique, introduced by the author, for high-speed digital acquisition and processing with maximum efficiency, that is using the minimum number of sampled signal data for the required processing, is described. This technique using a single digital filter and a frequency shift procedure of the sampled signal spectrum, is particularly useful for performing a band-pass analysis or a spectral estimation with a tree-structure organization. Hardware and software implementationsof the above technique are presented with typical application examples.