I always watch your videos they honestly amazing and the way you demonstrate is so great, however I am looking forward to watching anything you could do it about FBMC especially its channel estimation
I'm so glad you like the channel. And thanks for the topic suggestion. I'll add it to my "to do" list. In the mean time, you might like to watch the following video, which explains the effect of the pulse shaping (D-to-A) filter in OFDM, and shows that if you choose the filter (g(t)) to be a sinc function, then you can get rid of the sidelobes in OFDM - which is basically just what FBMC is doing: "How are OFDM Sub Carrier Spacing and Time Samples Related?" ruclips.net/video/knjeXo3VZvc/видео.html
You said when there are 2 frequencies, theiy could canceling out. I just know that canceling from the time domain - but that is here also possible in the frequency domain? Do you mean after the convolution of the rectangular and ther 2-frequencies-containing-pattern in the freq.-domain?
Actually, I didn't say "cancelled out", I said "swamped". What I meant was, that if the amplitude of the second frequency component is smaller than the amplitude of the side-lobe of the main frequency component, then it will be hard to decide whether the second frequency component is actually in the signal, or not.
Thank you for these great videos. How are these windows and their duration chosen when one examines actual occupied band of a communication system? If we are given 20Mhz bandwidth for 5g, and have a freedom of choosing the window duration, and signal is dynamic in time, how is judgment about actual physical presence of spectrum at a given time made?
When signals are random (or "dynamic in time" as you put it), we look at the "power spectral density". It can be thought of as an 'average Fourier transform energy". Here's a video on that: "What is Power Spectral Density (PSD)?" ruclips.net/video/DoSLMEEo1Y0/видео.html
I always watch your videos they honestly amazing and the way you demonstrate is so great, however I am looking forward to watching anything you could do it about FBMC especially its channel estimation
I'm so glad you like the channel. And thanks for the topic suggestion. I'll add it to my "to do" list. In the mean time, you might like to watch the following video, which explains the effect of the pulse shaping (D-to-A) filter in OFDM, and shows that if you choose the filter (g(t)) to be a sinc function, then you can get rid of the sidelobes in OFDM - which is basically just what FBMC is doing: "How are OFDM Sub Carrier Spacing and Time Samples Related?" ruclips.net/video/knjeXo3VZvc/видео.html
Always enjoy watching your videos Iain... learn something every time. Rgds, Howard W.
Glad to hear it! Thanks for letting me know. It's always great to hear from viewers who like the content.
You said when there are 2 frequencies, theiy could canceling out. I just know that canceling from the time domain - but that is here also possible in the frequency domain?
Do you mean after the convolution of the rectangular and ther 2-frequencies-containing-pattern in the freq.-domain?
Actually, I didn't say "cancelled out", I said "swamped". What I meant was, that if the amplitude of the second frequency component is smaller than the amplitude of the side-lobe of the main frequency component, then it will be hard to decide whether the second frequency component is actually in the signal, or not.
Thank you for these great videos. How are these windows and their duration chosen when one examines actual occupied band of a communication system? If we are given 20Mhz bandwidth for 5g, and have a freedom of choosing the window duration, and signal is dynamic in time, how is judgment about actual physical presence of spectrum at a given time made?
When signals are random (or "dynamic in time" as you put it), we look at the "power spectral density". It can be thought of as an 'average Fourier transform energy". Here's a video on that: "What is Power Spectral Density (PSD)?" ruclips.net/video/DoSLMEEo1Y0/видео.html