Is there an error in the "Power Spectral Density Estimates Using FFT" article?
조회 수: 1 (최근 30일)
이전 댓글 표시
In the article "Power Spectral Density Estimates Using FFT" found at https://www.mathworks.com/help/signal/ug/power-spectral-density-estimates-using-fft.html, should the line that says: 'psdx = (1/(Fs*N)) * abs(xdft).^2;' instead say: 'psdx = 1/Fs*N * abs(xdft).^2;' ? This would make it consistent with the "Power Spectum Estimate" block found inside the "doc_phasenoise" Simulink model that is opened by typing "doc_phasenoise" at the MATLAB command line. (See http://www.mathworks.com/help/releases/R2016a/comm/ug/rf-impairments.html )
댓글 수: 0
답변 (1개)
Kushagr Gupta
2016년 12월 19일
I understand that the question is regarding whether the variable 'N' should be in the numerator or the denominator of the power spectral density calculation.
The equation (4) in this link explains the relation between PSD and FFT, which points out that N is in the denominator.
Also, the model 'doc_phasenoise' contains N in the denominator inside the "spectral averaging" subsystem.
Hence, I believe it is right to say:
psdx = (1/(Fs*N)) * abs(xdft).^2;
댓글 수: 2
참고 항목
카테고리
Help Center 및 File Exchange에서 Spectral Estimation에 대해 자세히 알아보기
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!