Jeez this just made my day. It was so easy to easy to understand this way. You just want a periodic signal. That is the whole goal, with windowing that FTT and bam you have your awnser :0
I'm going to have critique this a little bit. Spectral leakage has nothing to due with one's ability to perfectly sample a full period of a signal or 'discontinuities' as your video suggests. The fact that you're windowing the signal is what induces the leakage. If you look at the DTFT of both of your cases (perfect period vs non-perfect period), you will see the windowing regardless. It's the fact that the DFT is a sampled version of the DTFT is what makes a difference. When you sample a perfect full period, the DFT sample points lines up with the lobes in the DTFT such that the leakage become invisible.
Hello, I really like your video very much and the explanation helped a lot. However, when you showed the window at around 6:42, the window function seemed to be misplaced on the axis. I wonder if I am seeing it right :)
Hi there, great video. I was curious to know , with a synchronous measurement technique, where the no. of samples captured is dependent on the shaft revolution, would there still be chances of spectral leakage?? Rather can a synchronous measurement be termed as a periodic measurement???
If the captured time signal when appended to be infinitely long, doesn't have discontinuities, then there is no need for windowing as signal is already periodic. In reality, this is rarely the case. Windowing is thus a default step in signal processing before performing FFT. By definition, a periodic motion is one that repeats itself in equal intervals of time. If any motion exhibits this phenomenon, then it can be considered periodic.
Jeez this just made my day. It was so easy to easy to understand this way. You just want a periodic signal. That is the whole goal, with windowing that FTT and bam you have your awnser :0
This is a fantastic introductory video to windowing!
fantastic video, better explanation provided than my prof. did. Bro is professional
Thank you so much for this video. Sincerely, your videos make signal processing so simple and interesting. Thank you Brother.
this is so helpful, accurate and easy to understand
I'm going to have critique this a little bit. Spectral leakage has nothing to due with one's ability to perfectly sample a full period of a signal or 'discontinuities' as your video suggests. The fact that you're windowing the signal is what induces the leakage. If you look at the DTFT of both of your cases (perfect period vs non-perfect period), you will see the windowing regardless. It's the fact that the DFT is a sampled version of the DTFT is what makes a difference. When you sample a perfect full period, the DFT sample points lines up with the lobes in the DTFT such that the leakage become invisible.
Thanks for your videos!
Keep uploading, please!!!
amazing video
Too cool, you do a great job, I can understand. keep up, thank you
Damn, that was an amazing explanation. Thanks a lot.
great video
Amazing, thx! :)
Great video, thank you!
Thanks.
Hello, I really like your video very much and the explanation helped a lot.
However, when you showed the window at around 6:42, the window function seemed to be misplaced on the axis. I wonder if I am seeing it right :)
Thank you. The horizontal axis for window function should rather be lower, please ignore it.
Great video!
Can i replace window function with a low pass filter?
Hi there, great video. I was curious to know , with a synchronous measurement technique, where the no. of samples captured is dependent on the shaft revolution, would there still be chances of spectral leakage?? Rather can a synchronous measurement be termed as a periodic measurement???
If the captured time signal when appended to be infinitely long, doesn't have discontinuities, then there is no need for windowing as signal is already periodic. In reality, this is rarely the case. Windowing is thus a default step in signal processing before performing FFT.
By definition, a periodic motion is one that repeats itself in equal intervals of time. If any motion exhibits this phenomenon, then it can be considered periodic.
thank you