Article ID Journal Published Year Pages File Type
7211294 Alexandria Engineering Journal 2016 13 Pages PDF
Abstract
A discrete resonant band-pass filter with a fast calculation algorithm, which can be used to perform discrete frequency transformations, is presented. The algorithm has low memory consumption requirements. It implements a numerical integration method, simulating a harmonic resonator element modeled by the under-damped driven oscillator equations, expressed in a discrete form. The output from the presented filter is a discrete function with an amplitude of the steady-solution that closely matches the theoretical steady-solution amplitude of the continuous band-pass filter output. Multiple discrete resonant band-pass filters can be used to build a filter bank, which in turn can be used to perform a time-to-frequency transformation of discrete signals. The filter achieves a frequency and a time localization without utilizing the time windowing method. The presented stand-alone calculation algorithm related to this filter produces its output with a delay of just one sampling period. The algorithm's calculation cost is only 3 multiplications and 3 additions per sample, and does not require long memory buffers. The presented transformation does not surpass the precision of the Discrete Fourier and Discrete Wavelet Transformations. However, it may prove essential when the noise-artifacts of the near-real-world simulation are necessary in order to produce some specific auditory-perception phenomena.
Related Topics
Physical Sciences and Engineering Engineering Engineering (General)
Authors
,