Effects of ripple width on ripple density resolution
Alexander Ya. Supin, Olga N. Milekhina, Dmitry Nechaev, Marina Tomozova
Abstract
Alexander Ya. Supin, Olga N. Milekhina, Dmitry Nechaev, Marina Tomozova
Abstract
Effects of ripple width in rippled-spectrum signals on ripple density resolution was investigated. Two measurement paradigms were tested: (i) ripple density resolution for discrimination between two rippled signals and (ii) discrimination between a rippled test signal and non-rippled reference signal. The ripple widths varied from 9% to 64% of the ripple frequency spacing. For both paradigms, the ripple density resolution increased with deceasing the ripple width. For discrimination between two rippled signals, the resolution was 8.1 ripples/oct for a ripple width of 64% and increased to 15.1 ripples/oct at the ripple width of 9%. For discrimination between a rippled test and non-rippled reference signal, the resolution was 9.3 ripples/oct at a ripple width of 64% and increased to 85 ripples/oct at a ripple width of 9%. Discrimination between two rippled signals is hypothesized to depend on ripple depth in the excitation pattern; the depth increases with narrowing the ripple width. Discrimination between a rippled test and non-rippled reference signal is hypothesized to depend on temporal processing; the effect of the ripple width appears due to increasing the ratio of the autocorrelated to uncorrelated components of the input signal with narrowing the ripples.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Effects of ripple width in rippled-spectrum signals on ripple density resolution was investigated. Two measurement paradigms were tested: (i) ripple density resolution for discrimination between two rippled signals and (ii) discrimination between a rippled test signal and non-rippled reference signal. The ripple widths varied from 9% to 64% of the ripple frequency spacing. For both paradigms, the ripple density resolution increased with deceasing the ripple width. For discrimination between two rippled signals, the resolution was 8.1 ripples/oct for a ripple width of 64% and increased to 15.1 ripples/oct at the ripple width of 9%. For discrimination between a rippled test and non-rippled reference signal, the resolution was 9.3 ripples/oct at a ripple width of 64% and increased to 85 ripples/oct at a ripple width of 9%. Discrimination between two rippled signals is hypothesized to depend on ripple depth in the excitation pattern; the depth increases with narrowing the ripple width. Discrimination between a rippled test and non-rippled reference signal is hypothesized to depend on temporal processing; the effect of the ripple width appears due to increasing the ratio of the autocorrelated to uncorrelated components of the input signal with narrowing the ripples.
Key concepts: Ripple, Resolution (logic), Optics, SIGNAL (programming language), Acoustics, Autocorrelation, Uncorrelated, Physics