基于小波变换的特征提取脑诱发电位 论文.docVIP

大 连 民 族 学 院 本 科 毕 业 设 计（论 文） 快速提取诱发脑电算法的研究 学 院（系）：信息与通信工程学院通信工程专业Abstract There are many kinds of artifacts in the raw brain signals from scalp, such as eyes blinks, Electrocardiograph, electromyography and other mechanical noises, which could degenerate the real evoked potentials(EP). How to extract the underlying evoked potentials from noisy acquired data has became an important and urgent problem to be resolved. This paper describes two kinds of new rapid extraction algorithm Evoked Potentials – wavelet analysis and independent component analysis. Wavelet analysis theory is a new theory of signal process and it has good localization in both frequency and time do-mains. It makes the wavelet analysis suitable for time-frequency analysis. Using wavelet methods in de-noising, is an important aspect in the application of wavelet analysis. The fundamental principles of Independent Component Analysis (ICA) is through analysis of high-ranking statistical correlation between multidimensional observation data, find mutually independent implicit message content, complete removal of a high redundancy and independent sources extraction letter. This paper first introduced the principle of wavelet analysis and independent component analysis algorithm in detail, also explained the two methods applied in signal processing. second, use the wavelet threshold method and the independent component analysis algorithm FastICA EP on a pure noise signal added after denoising, the example verify the theoretical practical effect, also confirmed the reliability of the theoretical. Key Words：EP；wavelet analysis；noise rejection；Independent Component Analysis 目 录 摘 要 I Abstract II 1 绪论 1 1.1 课题的背景与研究意义 1 1.2本课题的发展状况 1 1.3本课题的主要内容 2 2 基于小波变换去噪研究 3 2.1 小波变换 3 2.1.1 连续小波变换 4 2.1.2 离散小波变换 5 2.2 小波阈值去噪概述 5 2.3 小波阈值去噪方法 6 2.4小波阈值去噪仿真 7 2.5小结 8 3 基于独立分量分析的去噪研究 10 3.1 独立分量分析 10 3.2 ICA定点算法的实现 10 3.2.1 数据的预处理 11 3.2.2 ICA固定点算法 12 3.3 固定点ICA算法的程序实现和仿真 14 3.4 小结 16 结 论 17 参 考