Applied DSP No. 4: Sampling and Aliasing

Applied Digital Signal Processing at Drexel University: In this video, I discuss the unintended consequences of sampling, aliasing.

== Errata ==
Starting around 12:30, the amount of audio aliasing is much greater than indicated in the animation, due to a bug in my resampling code. I'm currently investigating and hope to post a link to a corrected video soon. My thanks to ytubeleo for the catch!

==

I'm teaching the course again this Fall (September 2022), so I *really* will be posting more in this series in the coming weeks. Now would be a great time to subscribe to my channel, @Youngmoo Kim , for updates. Please leave feedback in the comments and let me know if there are DSP topics you'd like to see in future videos!

[hahahahahasjovt]

This is honestly the best explained series I have yet to see on youtube. Good job!

[NineInchFailz]

dude that aliasing visualization at 11:55 blew me away. It all made sense in that instant. It's the low end being reintroduced from the high end of the spectrum due to the negative frequency components. That taught me what 2 courses in signals/systems classes at university couldnt

[yabool]

Listen at 00:10:25 "These higher frequency sinusoid do exist they come into existence as soon as we sample.". Super!

[shdyo]

DSP exam in less than 3 hours. This series is an amazing recap of my course. Thank you

[nepomukullmann]

I have recently taken a uni class on this and I must commend the way you explain! Also one of if not the best intro on all of youtube haha amazing

[justinwong]

Great series! Really helpful and easy to understand for a beginner.

[StephanBuchin]

Great work and excellent videos, clear and concise. Thx ☺

[ytubeleo]

At 12:15, I think the sampling rate numbers must be a bit out of sync with what we are hearing because there is obvious aliasing even when it says 44.1 kHz, and the aliasing is already bad at 40 kHz. If CD audio has no frequencies above 20 kHz (which it shouldn't by standard), then there shouldn't be any aliasing at all if we sample it at 40 kHz. And even at 36 kHz, a 20 kHz sine wave should reflect off the 18 kHz Nyquist limit and become 16 kHz, which is extremely high frequency and almost inaudible for most adults. Yet the aliasing sounds far lower in frequency. Or maybe something is going wrong when you convert it back into 44.1 kHz (or 48 kHz) for YouTube. What is going on here?

[rogerfederer]

please keep going. this stuff is gold.

[BroAlex]

Awesome series! Thank you for your work.

[svetoslavasenov]

Thank you! Where can I find #5 of the series? It is missing from your channel.

[abhijithas]

If sampling frequency increase after nyquist frequency then also aliasing occurs right ?
Is higher frequency component toke the place of lower frequency component or not ????
Is nyquist frequency is sampling frequency/2

Then sampling frequency= 2* highest frequency component of the signal or >=2 * highest frequency component of the signal

[mirosaide]

How can one knows the highest frequency component without doing an FFT?

[RooftopDuvet]

Really well paced and animated. Well done, and thanks!