I have a signal runs for 500 seconds ( 1:1:500 sec) after i did the fft, i want to go back to time domain and see how the signal looks like for 1000 Seconds (1:1:1000) ? Any advice how to extend the time ? Is it better to do it in frequency domain ? Is it simply to add the sines and cosines with loops ?
Thank you

 채택된 답변

Vilnis Liepins
Vilnis Liepins 2013년 11월 11일

0 개 추천

ifft(edft(YourData,1500),1500)
Signal for 1000 sec will be ifft output 1:1000.

댓글 수: 2

omar thamer
omar thamer 2013년 11월 12일
Thank you Vilnis, really helped a lot.Though, results not as expected for real signal but absolutely true for periodic computer generated one. I am wondering if you can advice on my signal which i attached here. As you can see a lot o discontinuities and no clear center frequency.
Any advice how to get periodicity to my signal ? I tried with STFT it also yields no results.
Its not working properly .. Have a look to my code:
t=1:1:400;
x=sin(2*pi*t*1/20);
X=edft(x,500);
Y=ifft(X,500);
the outcome for extended signal has different magnitude and frequency.

댓글을 달려면 로그인하십시오.

추가 답변 (1개)

Walter Roberson
Walter Roberson 2013년 11월 8일

0 개 추천

Try
ifft(fft(YourData), 1000)

댓글 수: 2

omar thamer
omar thamer 2013년 11월 8일
it will be padded with zeros then and return a complex vector. Its not working
Walter Roberson
Walter Roberson 2013년 11월 11일
fft() assumes that its data is periodic, so to get 1000 seconds of output, ifft() to get the 500 second output and then replicate it.

댓글을 달려면 로그인하십시오.

카테고리

도움말 센터File Exchange에서 Fourier Analysis and Filtering에 대해 자세히 알아보기

태그

질문:

2013년 11월 8일

댓글:

2013년 11월 13일

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by