期刊名称：International Journal of Signal Processing, Image Processing and Pattern Recognition
印刷版ISSN：2005-4254
出版年度：2015
卷号：8
期号：8
页码：135-142
DOI：10.14257/ijsip.2015.8.8.15
出版社：SERSC
摘要：Power spectral density (PSD) is an effective way to analyze random signals. However, it is a time-consuming calculation when it deals with a rapidly growing massive data by using the serial Matlab circumstances. Though a parallel computing toolbox is involved in Matlab 2004, it is too expensive to be wildly used. According to an algorithm of PSD, a parallel algorithm of cross power spectral density (PACPSD), which was implemented using the Master-Slave parallel programming model with the support of Linux clusters and MPI ， was proposed here based on the theory of Welch algorithm. The experiment results reveal that PACPSD can achieve the same accuracy as the function cpsd in Matlab do, and greatly reduce the operation cost
关键词：cross power spectral density; Welch algorithm; message passing interface
