摘要
本文首先提出了客观评价噪声压缩效果的两个指标：残余噪声能量和信号畸变指标。并在此基础上提出了一种基于主分量分析的时间序列噪声压缩方法─噪声压缩链。这种方法充分利用了信号和白噪声之间的统计特征差别，能够有效地压缩时间序列中的白噪声，同时使序列中的信号畸变较小。实例分析结果说明了该方法的有效性。
This paper firstly presents two criteria: residual noise energy and signal distortion criterion to evaluate the degrees of noise suppression and signal distortion. Next, on the basis of the two criteria, a numerical algorithm based on Principal Component Analysis (PCA)-Noise Suppression Chain is proposed. The method makes full use of the statistical characteristics of signal and noise, it can suppres the major part of noisecontaimed in the signal and retain chief charateristics of the original signal as much as possible. Finally, the examples are given to verify its effectiveness.
出处
《信号处理》
CSCD
1998年第4期318-324,共7页
Journal of Signal Processing
关键词
主分量分析
噪声压缩
信噪比
信号统计
filtering, principal component analysis, nonlinear dynamics
