基于高阶谱的肌纤维传导速度谱坑法检测

Spectrum Dip Estimator of Muscle Fiber Conduction Velocity Based on High-order Spectra

针对目前采用的谱坑法抗噪声性能差的问题 ,提出了基于高阶谱重构功率谱的谱坑法 ,检测肌纤维传导速度 (MFCV) ,并对这种方法的抗噪性能进行了研究。同时 ,提出一种基于小波多尺度分析的适于噪声环境下谱坑自动检测的多尺度 -二阶差分法 ,完成对肌纤维传导速度的检测。仿真实验结果表明 ,本文的方法与经典周期图估计功率谱谱坑法相比 ,可以显著改善估计功率谱的信噪比 ,提高谱坑法的抗噪能力和MFCV检测的精度。对真实针电极肌电信号的实验结果也表明 。

A new spectrum dip estimator of muscle fiber conduction velocity (MFCV) based on the reconstructed power spectrum using high order spectra estimation of EMG signal was proposed and the anti noise ability of the method was investigated. The wavelet multiscale second order differential algorithm based on wavelet multi resolution analysis ,which is suitable for spectrum dip detection in noise environment, was proposed at the same time. Simulation experiments showed that the method could improve the signal noise ratio of the power spectrum estimation distinctly compared to those estimated by classical periodogram method, and could enhance the noise resistance ability and the precision of MFCV measurement. The analysis of real needle electrode EMG signal also showed that the value of MFCV measured by the method was in accordance with the value range given by normal methods.