基于高阶统计量分析的生物医学信号处理应用

摘要

目的:高阶统计量是随机过程的新的数字特征,通常是指高阶矩、高阶累积量以及它们的谱--高阶矩谱和高阶累积量谱.许多生物医学信号具有非线性、非平稳、非高斯、非确定性的固有特征.高阶统计量分析在信号处理某些应用中能提供特征信息量,具有一定的优势.本文探讨高阶统计量分析在生物医学信号处理的应用.方法:介绍了有关高阶矩、高阶累积量、高阶累积量谱的理论基础,阐述了高阶统计量用于处理脑电、心电、表面肌电信号、肺音、心音、二维生物医学信号等信号,提取出有用信息的应用.结果:高阶统计量不仅包含原信号的幅度信息,还包含其相位信息,能解决非高斯、非线性问题,在理论上可以完全抑制高斯有色噪声的影响,是一种分析具有非线性特征的生物医学信号的理想工具.结论:高阶统计量分析具有非线性、非平稳、非高斯、非确定性的生物医学信号具有重要的应用价值.%Objective: Higher order statistics are new features of the random digital signal processing, usually refers to higher order moments, higher order cumulants and higher order spectra including higher order moment spectrum and higher order cumu-lant spectrum. To discuss the applications of higher order statistics analysis in biomedical signal processing. Methods: The basic principles of higher order moments, higher order cumulants and higher order spectra are introduced in this article and the applications of higher order statistics for different bio-signals such as EEG, ECG, EMG, respiratory, heart sounds and the two-dimensional biological medical signals are discussed. Results: Higher order statistics processes contains both amplitude and phase information of the original signal, which can solve the non-Gaussian, nonlinear problem, in theory, and can completely abolish the Gaussian colored noise. Conclusions: Higher order statistics can deal with nonlinear, non-stationary, non-Gaussian, non-deterministic biomedical signals and is powerful in biomedical signal processing.