论文部分内容阅读
压缩感知CS(Compressive Sensing)作为一门新兴的技术,成为近年来人们广泛关注的研究热点。文中介绍了压缩感知理论的基本原理,在此基础上将压缩感知理论应用到语音信号处理中。首先研究了语音信号的稀疏性,说明了对语音信号进行压缩感知具有可行性;其次,采用随机滤波器组构造随机测量矩阵得到语音信号的压缩测量值;最后,研究了压缩测量值之间的相关性并将这种相关性作为稀疏度的一种度量方法用于控制随机滤波器阶数,实现了语音信号的自适应压缩感知。
As a new technology, Compressive Sensing (CS) becomes a research hotspot that has attracted a great deal of attention in recent years. In this paper, we introduce the basic theory of compressed sensing theory, and apply compressed sensing theory to speech signal processing. Firstly, the sparseness of speech signal is studied, which shows that it is feasible to compress the speech signal. Secondly, a random measurement matrix is constructed by using a random filter to obtain the compression measurement of speech signal. Finally, Correlation and use this correlation as a measure of sparsity to control the order of random filters and achieve adaptive compression perception of speech signals.