Very well explained, the heart and soul of the FIR filter, ie, it's underlying nature and purpose. So many other videos went through so much talk without getting to the very bottom of simply WHAT an FIR filter is; and what it's primarily used for.
Great Explanation! I really like the way you related the FIR filter to convolution and reiterating convolution is the same as multiplication in the frequency domain, really so simple but I never truly made a complete comprehension of this. So 11 minutes well spent. Thanks.
what determine the sampling frequency of the input x(t)? Is there a way to increase the sampling frequency of x(t) ? And does the increased sampling frequency affect the Z delay (increased sampling frequency will make Z delay shorter)?
Thanks for the video. My question pertains the cut-off filtering frequency: When you apply a HP filter in fMRI (typically it's .008 Hz), is this at the Nyquist frequency? Meaning that .008/2 is the actual frequency we are removing? Thank you.
I'm no expert on MRIs, but something in your description isn't right. HP = high-pass filter, which passes frequencies *above* the cutoff. You would not use a high-pass filter in relation to the Nyquist frequency unless you were purposely operating in an aliased Nyquist band (which isn't for the faint of heart). Moreover, filters are always (AFAIK) at the frequency they specify, so a HP 0.008 Hz filter would have the low-frequency cutoff at 0.008 Hz.
hi arron. if you are reading this,. first let me tell you this. your videos are really good but your channel is tremendously underrated. I tell you the reason. Most of filter videos i find in the internet were complete scrap taught by people who themselves didn't seem to realize what they were talking about.but those channels had views because they taught according to college syllabus. even if they are total junk, they the the best shot students have to read something in internet. So, coming to the main point, . let me tell to something to hit your channel. the most precious viewers for your channel will be students. Make a signal processing playlist that gives both intuition (as here) along with target according to syllabus. look at any major university course for digital signal processing. solve some numerical problems for fourier and other signal processing stuffs as well. i promise, within a year, your subscribes will spike up (considering those are high quality content as above)
Generally well done, but what I'm missing here: 1. Difference to IIR Filters and how to decide which one to use for which case and why 2. Usually the filter is represented as transfer function (in digital (z) or laplace (s) domain), here you have f(t), which doesnt make sense to me since a filter is NOT a timeseries of numbers. 3. (Dis)advantages of FIR Filters 4. While it makes sense that this filter has a finite impulse response when looking at the block diagram, a short explanation would have been nice Greetings
fir is nothing's just a transistors used to sample a sample frequency ... double the frequency is needed for a frequency so no frequency will miss when the transistors doesnot off while analog frequency is on.
It is misleading to talk about FFT (Fast Fourier Transform) when what is actually relevant here is a straightforward DFT (Discrete Fourier Transform). The "F" or "Fast" aspect of the FFT algorithm is neither touched on here nor necessary.
I agree that the "Fast" is irrelevant, but FFT has also become common parlance for numerically computing a Fourier Transform. I think distinguishing between "D" and "F" in this video might be overly pedantic.
Very well explained, the heart and soul of the FIR filter, ie, it's underlying nature and purpose. So many other videos went through so much talk without getting to the very bottom of simply WHAT an FIR filter is; and what it's primarily used for.
Great Explanation! I really like the way you related the FIR filter to convolution and reiterating convolution is the same as multiplication in the frequency domain, really so simple but I never truly made a complete comprehension of this. So 11 minutes well spent. Thanks.
Helped me a lot to understand FPGA design. Don't know why I go to school, internet is way better
Then from now stop going to school.
@@n_3719 you are two months late my friend....oh wait!
best FIR filter explanation i've seen on youtube
"To implement the filter of your dreams". I really appreciated that
Okay, now I understand why having an efficient Multiply-and-accumulate instruction is important in digital signal processing.
great video! a lot of profs have hours of videos online, but no one understands what they are trying to say...
2mins in and I already love this. Subscribed.
I think I understand it now. thanx for the pretty clear explanation.
You explained this very very well! THANK YOU
This was some great information fir sure!
hi ... multiplying by coefficient f0 f1 f2 f3 ,, what are they where and how do we calculate them ... what do they represent ? thx
That is inverse Fourier transfrom of the filter back to become discrete time impulse response of the filter to be applied.
when we add delay on x or input, do we need to flip the input signal like in convolution, I am confused.
I don't see any FIR filter in YT mentioning to phase response of the filter. Only magnitude is stressing.
Thanks for the great introduction to filters!
what determine the sampling frequency of the input x(t)? Is there a way to increase the sampling frequency of x(t) ? And does the increased sampling frequency affect the Z delay (increased sampling frequency will make Z delay shorter)?
Can you please give an example of a generalized linear-phase lter which is not FIR, and does not exhibits odd or even symmetry?
Very very elegant explanation
Thanks for the video. My question pertains the cut-off filtering frequency: When you apply a HP filter in fMRI (typically it's .008 Hz), is this at the Nyquist frequency? Meaning that .008/2 is the actual frequency we are removing? Thank you.
I'm no expert on MRIs, but something in your description isn't right. HP = high-pass filter, which passes frequencies *above* the cutoff. You would not use a high-pass filter in relation to the Nyquist frequency unless you were purposely operating in an aliased Nyquist band (which isn't for the faint of heart). Moreover, filters are always (AFAIK) at the frequency they specify, so a HP 0.008 Hz filter would have the low-frequency cutoff at 0.008 Hz.
Good explanation Aaron, thanks!
hi arron. if you are reading this,. first let me tell you this. your videos are really good but your channel is tremendously underrated. I tell you the reason. Most of filter videos i find in the internet were complete scrap taught by people who themselves didn't seem to realize what they were talking about.but those channels had views because they taught according to college syllabus. even if they are total junk, they the the best shot students have to read something in internet.
So, coming to the main point,
. let me tell to something to hit your channel. the most precious viewers for your channel will be students. Make a signal processing playlist that gives both intuition (as here) along with target according to syllabus. look at any major university course for digital signal processing. solve some numerical problems for fourier and other signal processing stuffs as well. i promise, within a year, your subscribes will spike up (considering those are high quality content as above)
This is the best explanation of FIR Filters I've found. Question: are the number of taps in an FIR filter related to the FFT size?
Thank you, it was very helpful to get an overall idea ! :) Cheers.
Generally well done, but what I'm missing here:
1. Difference to IIR Filters and how to decide which one to use for which case and why
2. Usually the filter is represented as transfer function (in digital (z) or laplace (s) domain), here you have f(t), which doesnt make sense to me since a filter is NOT a timeseries of numbers.
3. (Dis)advantages of FIR Filters
4. While it makes sense that this filter has a finite impulse response when looking at the block diagram, a short explanation would have been nice
Greetings
thank you so much !! you helped me a lot !!
perfecto, muchos gracias amigo
Thank you so much. Perfect video :)
Well explained...!!!
Very good video, thanks a lot
Hi great work. Do you Code? I have works to do. Can you help me? we can talk about the work and the payment.
uoooh this guy magic!!!
Thank god I have an audio card that amplifies the volume...
can we filter 45 dislikes from 580[dis/like] and slap them?
Any literature/book recommendations? Would be great. Thanks!
"... the filter of your dreams..."
/Sensible Chuckle
fir is nothing's just a transistors used to sample a sample frequency ... double the frequency is needed for a frequency so no frequency will miss when the transistors doesnot off while analog frequency is on.
FIR是一種有效的方法來實作卷積運算
Vioce audiable is low
But what does it dooo!!!???
it filters dude :D
It helps to narrow down our focus to only a certain frequency component of the original signal
Audio is very low.
Volume is too low
It is actually x(t).f(t). ::::: X(e^jw)*F(e^jw)
At 3:19
It is misleading to talk about FFT (Fast Fourier Transform) when what is actually relevant here is a straightforward DFT (Discrete Fourier Transform). The "F" or "Fast" aspect of the FFT algorithm is neither touched on here nor necessary.
I agree that the "Fast" is irrelevant, but FFT has also become common parlance for numerically computing a Fourier Transform. I think distinguishing between "D" and "F" in this video might be overly pedantic.
You should leave teaching about FIR filters to someone who thinks properly about the words they use.
MrCuddlyable3 FFT is just an algorithm for computers for calculating the DFT...
hemnexus7 why do you find it necessary to repeat that information?
Caleb Geballe so are you here to entertain us?
u sounds like n0thing lel
so bad sound quality.wasted of time