诱发电位提取的聚类分析和小波去噪复合方法

Visual Evoked Potential (VEP) Estimation, Using a Method Combining Clustering Algorithm and Wavelet Noise Reduction Technigues

本文研究了提取诱发电位的一种复合方法 ,它用聚类方法对诱发电位信号进行筛选 ,将筛选出的信号进行叠加以消除测试数据不一致性对叠加结果的影响。然后利用小波方法进一步去除噪声 ,提高信噪比。文中介绍了模糊聚类方法和小波去噪理论 ,并通过仿真计算来评估提取效果。仿真计算表明该方法可减少测试次数 ,提高信噪比 ,对视觉脑干诱发电位处理结果显示 ,该方法的实际处理效果良好。

A combination method to estimate VEP is presented in this paper. In the method, the fuzzy clustering algorithm was used to select a group of samples with similar characteristics. Signals were averaged to eliminate influence of non uniformity of test data on results of superimposition. Then wavelet method was used to further reduce the noise. Fuzzy clustering method and the theory of eliminating noise with wavelet are introduced in this paper. Simulation calculation was carried out to evaluate effects of extraction. Simulation calculation demonstrated that this method could effectively reduce the number of times of measurement, and could increase the signal to noise ratio. The results of processing visual brain stem evoked potential showed that the practical processed results were good.