ICA不确定性问题在圆度误差分离中的解决措施

Solving Method of ICA Ambiguity Problem in Roundness Error Separation

独立分量分析方法(Independentcomponentanalysis,简称ICA)在国内尚属一门新型的方法。文章介绍了ICA的无噪声模型、原理、预处理、非高斯性量度以及快速定点算法,重点讨论了ICA的不确定性在圆度误差分离中的处理方法。仿真结果表明,基于独立分量分析的圆度误差分离技术比传统的频域法和时域法更简单、实用、高效,同时由ICA分离出的信号不确定性问题得到了很好的解决。

Independent component analysis （ICA） is a new method in abroad. The ICA noiseless model, principles, preprocessing, measures of non-Gaussianity and fast ICA algorithm are introduced. The method of processing the ambiguities of ICA in roundness error separation is emphasized. The emulation result shows that the roundness error separation technique （EST） based on ICA is simple, practical and efficient, which is better than the traditional frequency-domain and time-domain method. The problems of ambiguities are well solved.