원문정보
보안공학연구지원센터(IJSIP)
International Journal of Signal Processing, Image Processing and Pattern Recognition
Vol.8 No.8
2015.08
pp.135-142
피인용수 : 0건 (자료제공 : 네이버학술정보)
초록
영어
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.
목차
Abstract
1. Introduction
2. CPSD Algorithm and its Serial Implementation
2.1 CPSD Algorithm
2.2 Serial Implementation of the CPSD Algorithm
3. Parallel Algorithm of the CPSD
3.1 Program Structure of PAcpsd
3.2 Distribution of Tasks
4. Experimental Results and Analysis
4.1 Problem Description
4.2 Experimental Environment and Evaluation Standard
4.3 Analysis of Experimental Results
5. Conclusion
Acknowledgements
References
1. Introduction
2. CPSD Algorithm and its Serial Implementation
2.1 CPSD Algorithm
2.2 Serial Implementation of the CPSD Algorithm
3. Parallel Algorithm of the CPSD
3.1 Program Structure of PAcpsd
3.2 Distribution of Tasks
4. Experimental Results and Analysis
4.1 Problem Description
4.2 Experimental Environment and Evaluation Standard
4.3 Analysis of Experimental Results
5. Conclusion
Acknowledgements
References
저자정보
참고문헌
자료제공 : 네이버학술정보