A Time Domain Digital Weighting Method for Nonstationary Noise Signals

Abstract

The invention relates to a time-domain digital weighting method for non-stationary noise signals, which comprises the following steps: 1) collecting data on non-stationary noise signals to obtain noise digital signals p(τ); 2) according to the noise weighting network The octave sound level correction value constructs the weighted amplitude curve A(f); 3) performs mirror transformation processing on the weighted amplitude curve A(f), and obtains the aperiodic real even function A'( f): 4) Perform inverse Fourier transform to obtain the impulse response function A'(t) of the weighting network; 5) Select the window function Wβ(t) to perform windowing and truncation processing on A'(t); 6) Obtain the weighting network Window weighted wavelet function A'wβ(t); 7) Interpolate and resample the window weighted wavelet function so that its time interval is consistent with the time interval of the original noise digital signal collected; 8) The collected non-stationary The noise digital signal p(τ) is compared and transformed with the window weighted wavelet function to obtain the time domain weighted fluctuation signal pw(t); 9) The weighted total sound pressure level of the non-stationary signal is obtained by calculating the sound level transformation Curve Lw(t).

Description

Technical field [0001] The invention relates to a noise signal analysis and processing method, in particular to a time-domain digital weighting method for non-stationary noise signals whose frequency components change rapidly with time. Background technique [0002] The human ear has a large dynamic range of sound perception, and the loudness perceived by the hearing is proportional to the logarithm of the sound pressure ratio. Therefore, the concept of sound pressure level is used in acoustics to reflect the characteristics of the human ear’s self-perceived perception of loudness . In addition, due to the special structure of the human ear, the human ear has different sensitivities to noise of different frequencies. In order to obtain an evaluation standard consistent with human subjective perception, the measured decibel value has a certain degree of subjective loudness. Relevance, the International Organization for Standardization applies different frequency band weighting pr...

AI Technical Summary

Problems solved by technology

This calculation method is cumbersome and complicated, and requires multiple filtering calculations (the number of subbands), which requires a large amount of calculation. At the same time, there is also the problem of aliasing of subband spectrum energy and causing calculation errors.

In addition, this method cannot obtain the weighted time-domain fluctuation signal of the sound

Images