窗函数在扬声器异常声客观检测中的影响

The selection of window functions in loudspeaker Rub & Buzz detection

该文研究了基于短时傅里叶变换的扬声器异常声检测方法中,窗函数对特征距离阈值设定的影响。以连续对数扫频信号激励被测扬声器单元,并对响应信号进行短时傅里叶变换。通过对比不同窗函数下,被测单元之间的特征距离来寻找合适的窗函数。结果表明,不同窗函数下得到的特征距离具有不同的离散度;选择合适的窗函数,可使特征距离具有较大离散度,从而更合理地设定阈值。

This paper studies the influence of window function on the threshold setting of feature distancewhen using short-time Fourier transform STFT for detecting Rub ＆ Buzz of loudspeakers. The tested loudspeakersare excited by a continuous logarithm swept signal, and the measured sound signal is processed withSTFT. Different window functions are used and the appropriate window function is found by comparing thefeature distance between the measured loudspeakers. The results show that feature distances obtained by usingdifferent window functions have different dispersion degrees. An appropriate window function results in higherdispersion, which is beneficial for setting the threshold.