基于压缩感知理论的脑电信号压缩方法研究.doc
约35页DOC格式手机打开展开
基于压缩感知理论的脑电信号压缩方法研究,2万字自己原创的毕业论文，仅在本站独家出售，重复率低，推荐下载使用摘要 本文对课题“基于压缩感知理论的脑电信号压缩采样”进行了研究，首先简述了传统概念上的信号采样，同时也介绍了一种新的信号采样方法，即压缩感知理论，并对二者进行了对比。本文采用的方法为正交匹配追踪算法，并通过matl...
内容介绍
此文档由会员 淘宝大梦 发布
基于压缩感知理论的脑电信号压缩方法研究
2万字
自己原创的毕业论文，仅在本站独家出售，重复率低，推荐下载使用
摘要 本文对课题“基于压缩感知理论的脑电信号压缩采样”进行了研究，首先简述了传统概念上的信号采样，同时也介绍了一种新的信号采样方法，即压缩感知理论，并对二者进行了对比。本文采用的方法为正交匹配追踪算法，并通过Matlab编程对EEG采样信号的两组信号进行了仿真，并将最终复原的信号与原始信号进行了比对，效果较为理想。
在医学实践中，脑电信号的提取和采集复原是很重要的，但又因为脑电信号的信号数量繁多，因此会产生大量数据，给信号的采集与分析带来了很多不便。怎样高效采集分析这些数据是一个有待解决的问题。而有别于传统采样理论的压缩感知理论，为有效解决这个问题提出了全新的解决思路。基于这项全新的理论，本文首先介绍了EEG信号的基础知识以及传统的EEG的采样方法，然后介绍了压缩感知的理论框架，并用框图简单说明了这两种方法。
接下来，本文研究了基于压缩感知理论对单EEG信号的压缩采样，内容包括脑电信号最佳稀疏分解，通过实验对比，展示了EEG信号经过压缩感知理论采样还原后的效果。本文中主要采取的方法为正交匹配追踪法。通过实验我们可以实现EEG信号的较好的稀疏分解，在实验中，我们对测量矩阵进行了选择，比较了常用测量矩阵对重构误差的影响，接下来使用测量矩阵对稀疏分解系数向量进行观测得到测量值并完成压缩采样，最后由这些测量值运用正交匹配追踪算法恢复出系数向量，继而完成原EEG信号的重构。
关键词：压缩感知 压缩采样 EEG 正交匹配追踪法
Research on the Compression of EEG Based on Compressive Sensing
Abstract: This paper mainly does a research on the topic “The data acquisition of EEG based on compressive sampling”. Firstly， this paper introduces the traditional way of data acquisition，then a brand new theory which called “compressive sampling” is given. There are a few words used to compare the two ways of collecting signals and their pros and cons are shown as well.
In the medial practice，it is essential and complicated to collect enough and effective EEG signals. The huge number of the signals makes the collection process even harder and time-wasting. How to collect and analyze those signals with high efficiency has been a problem for a long time. Differed from the traditional ways，compressive sampling provides a new approach to that problem.
Based on that theory，this paper at first introduces the basic knowledge of EEG and old ways of collecting signals and then briefly introduces the outline of the compressive sampling with some flow diagrams for both methods. Next，there are some researches which were done to the EEG signals including its sparsity. Finally，the results of two methods are compared with each other.
This paper uses Orthogonal Matching Pursuit to reconstruct the original signal and through the experiment it can be seen that signal witch is collected and processed with that method turned to be recovered well. The matrix used in this article is common matrix and the testing matrix is made especially for the signals.
Through the compressive sampling we can collect the signals much smaller in the quantity and recover the signals with higher quality.
Key words: Compressive sensing EEG Orthogonal Matching Pursuit
目录
第一章 绪论 1
1.1研究背景、目的、意义 1
1.2 国内外相关领域的研究现状 4
1.3 EEG的基础知识 5
1.4 脑电压缩技术简介 9
1.5 压缩与重建测评指标 11
第二章 压缩感知理论 12
2.1传统采样与压缩感知的比较 12
2.2 压缩感知的模型架构简介 13
2.3 信号的稀疏表示 15
2.4 测量矩阵的设计 16
2.5 重构算法的简介 18
第三章 基于压缩感知理论的EEG采样 20
3.1 Matlab技术简介 20
3.2 稀疏基和冗余字典 21
3.3 正交匹配追踪算法 21
3.4 其他重构算法 22
3.5 正交匹配追踪算法流程图 23
第四章 总结与展望 26
致谢 27
参考文献 28
附录A 30
2万字
自己原创的毕业论文，仅在本站独家出售，重复率低，推荐下载使用
摘要 本文对课题“基于压缩感知理论的脑电信号压缩采样”进行了研究，首先简述了传统概念上的信号采样，同时也介绍了一种新的信号采样方法，即压缩感知理论，并对二者进行了对比。本文采用的方法为正交匹配追踪算法，并通过Matlab编程对EEG采样信号的两组信号进行了仿真，并将最终复原的信号与原始信号进行了比对，效果较为理想。
在医学实践中，脑电信号的提取和采集复原是很重要的，但又因为脑电信号的信号数量繁多，因此会产生大量数据，给信号的采集与分析带来了很多不便。怎样高效采集分析这些数据是一个有待解决的问题。而有别于传统采样理论的压缩感知理论，为有效解决这个问题提出了全新的解决思路。基于这项全新的理论，本文首先介绍了EEG信号的基础知识以及传统的EEG的采样方法，然后介绍了压缩感知的理论框架，并用框图简单说明了这两种方法。
接下来，本文研究了基于压缩感知理论对单EEG信号的压缩采样，内容包括脑电信号最佳稀疏分解，通过实验对比，展示了EEG信号经过压缩感知理论采样还原后的效果。本文中主要采取的方法为正交匹配追踪法。通过实验我们可以实现EEG信号的较好的稀疏分解，在实验中，我们对测量矩阵进行了选择，比较了常用测量矩阵对重构误差的影响，接下来使用测量矩阵对稀疏分解系数向量进行观测得到测量值并完成压缩采样，最后由这些测量值运用正交匹配追踪算法恢复出系数向量，继而完成原EEG信号的重构。
关键词：压缩感知 压缩采样 EEG 正交匹配追踪法
Research on the Compression of EEG Based on Compressive Sensing
Abstract: This paper mainly does a research on the topic “The data acquisition of EEG based on compressive sampling”. Firstly， this paper introduces the traditional way of data acquisition，then a brand new theory which called “compressive sampling” is given. There are a few words used to compare the two ways of collecting signals and their pros and cons are shown as well.
In the medial practice，it is essential and complicated to collect enough and effective EEG signals. The huge number of the signals makes the collection process even harder and time-wasting. How to collect and analyze those signals with high efficiency has been a problem for a long time. Differed from the traditional ways，compressive sampling provides a new approach to that problem.
Based on that theory，this paper at first introduces the basic knowledge of EEG and old ways of collecting signals and then briefly introduces the outline of the compressive sampling with some flow diagrams for both methods. Next，there are some researches which were done to the EEG signals including its sparsity. Finally，the results of two methods are compared with each other.
This paper uses Orthogonal Matching Pursuit to reconstruct the original signal and through the experiment it can be seen that signal witch is collected and processed with that method turned to be recovered well. The matrix used in this article is common matrix and the testing matrix is made especially for the signals.
Through the compressive sampling we can collect the signals much smaller in the quantity and recover the signals with higher quality.
Key words: Compressive sensing EEG Orthogonal Matching Pursuit
目录
第一章 绪论 1
1.1研究背景、目的、意义 1
1.2 国内外相关领域的研究现状 4
1.3 EEG的基础知识 5
1.4 脑电压缩技术简介 9
1.5 压缩与重建测评指标 11
第二章 压缩感知理论 12
2.1传统采样与压缩感知的比较 12
2.2 压缩感知的模型架构简介 13
2.3 信号的稀疏表示 15
2.4 测量矩阵的设计 16
2.5 重构算法的简介 18
第三章 基于压缩感知理论的EEG采样 20
3.1 Matlab技术简介 20
3.2 稀疏基和冗余字典 21
3.3 正交匹配追踪算法 21
3.4 其他重构算法 22
3.5 正交匹配追踪算法流程图 23
第四章 总结与展望 26
致谢 27
参考文献 28
附录A 30
