A nice continuation of this would be to show the spectrum of the modulated waveform so you can see that the bandwidth is half the bitrate for QPSK compared to BPSK, and that the bandwidth is determined by the symbol rate, not the bitrate (ruclips.net/video/ZBSvMbO0mPQ/видео.html). It would also be helpful to show the modulated waveform with and without a transmit filter in the frequency, time, and IQ polar domains to see how we can reduce the occupied bandwidth nearly for free as long as the receiver knows the transmit filter. That gets us to pi/2-BPSK and pi/4-QPSK, which allows us to avoid a zero crossing on the IQ constellation, which reduces the PAPR, nonlinear products when used in a real transmitter, and perhaps some improvements in the frequency domain.
Thank you for you videos. Do you have something on long training sequences? I woud like to mess a bit with wifi CSI, but I don't know exactly what I'm getting from the chipset
Sorry, I don't have any videos on that at the moment. What aspect are you wondering about? There's nothing particularly fundamental about long training sequences. They are generally generated from a sequence generator, so you need to know the starting state, and the start time.
@@iain_explains I found a video of yours where you talked a bout their use, that was enough, thank you. I was trying to understand what I should see in the CSI report of a wifi driver, and I suppose they removed the actual sequence, and left just the H matrix in the report.
Hello prof., i have i a problem and i wish you will help me to solve it. About cnr vs ber i calculate it using both simulation and real time device and i got the same result which it is 3 db different than the available reference.. As example for BPSK i got 1e-4 at 6 DB cnr but in the reference it must be 9 db atr this ber, i using noise single from matalb wgn function for both simulation and real-time and for more detail the signal is 1 vp that mean it is - 3 db and for 6 cnr i use - 9 db power noise.
One of the thing that I have been always wondering about is that how can we actually generate these sine wave and square wave. Is it really sine wave and square wave or just some "approximation" that look kinda sine and kind square.
The carrier which is a sine wave is not an approximate but a truly sine function generated using an analog local oscillator (LO). The same is true for the square function, although in practice we rather use roll-off cosine shaping to avoid out of band interference through the "sinc" spectrum of a pure square baseband signal.
idk how this guy keeps producing bangers
Glad you’re liking all the videos. 😁
Excellent video, very rare to see the wave form of qam
Thanks. I'm glad you liked it.
Thanks for the video! Any chance you could reveal the type of visualization software you have used?
excellent explanation
Glad you liked it!
Just catched up with one of your always. helpful videos. Greetings from Germany.
Glad you like them!
Best thumbnail ever (:
Fantastic video thank you very much
Glad you liked it.
A nice continuation of this would be to show the spectrum of the modulated waveform so you can see that the bandwidth is half the bitrate for QPSK compared to BPSK, and that the bandwidth is determined by the symbol rate, not the bitrate (ruclips.net/video/ZBSvMbO0mPQ/видео.html). It would also be helpful to show the modulated waveform with and without a transmit filter in the frequency, time, and IQ polar domains to see how we can reduce the occupied bandwidth nearly for free as long as the receiver knows the transmit filter. That gets us to pi/2-BPSK and pi/4-QPSK, which allows us to avoid a zero crossing on the IQ constellation, which reduces the PAPR, nonlinear products when used in a real transmitter, and perhaps some improvements in the frequency domain.
They're all great suggestions! I just need to find the time. In the meantime I've added them to my "to do" list. Thanks.
Well Explained. thank you.
Glad it was helpful!
Great job with this video and the channel Ian. I really appreciate it. Can you do a video about Probabilistic Amplitude Shaping (PAS)?
Thanks for the topic suggestion. I've added it to my "to do" list.
wow, this is once again amazing. Is this app / visualisation tool available somewhere?
Sorry, I'm still developing the visualisation tool.
Thank you for you videos. Do you have something on long training sequences? I woud like to mess a bit with wifi CSI, but I don't know exactly what I'm getting from the chipset
Sorry, I don't have any videos on that at the moment. What aspect are you wondering about? There's nothing particularly fundamental about long training sequences. They are generally generated from a sequence generator, so you need to know the starting state, and the start time.
@@iain_explains I found a video of yours where you talked a bout their use, that was enough, thank you. I was trying to understand what I should see in the CSI report of a wifi driver, and I suppose they removed the actual sequence, and left just the H matrix in the report.
💖💖💖💖
Hello prof., i have i a problem and i wish you will help me to solve it.
About cnr vs ber i calculate it using both simulation and real time device and i got the same result which it is 3 db different than the available reference..
As example for BPSK i got 1e-4 at 6 DB cnr but in the reference it must be 9 db atr this ber, i using noise single from matalb wgn function for both simulation and real-time
and for more detail the signal is 1 vp that mean it is - 3 db and for 6 cnr i use - 9 db power noise.
One of the thing that I have been always wondering about is that how can we actually generate these sine wave and square wave. Is it really sine wave and square wave or just some "approximation" that look kinda sine and kind square.
The carrier which is a sine wave is not an approximate but a truly sine function generated using an analog local oscillator (LO).
The same is true for the square function, although in practice we rather use roll-off cosine shaping to avoid out of band interference through the "sinc" spectrum of a pure square baseband signal.
@@BilelMnasri13 perfect square in time domain occupied infinite bandwidth to construct so is it some kind of approximately square looking pulse ?
@@tuongnguyen9391 Indeed, you are right. That is why we use roll-off cosine function instead of perfect square pulse shaping.
Here's a video on pulse shaping: "Pulse Shaping and Square Root Raised Cosine" ruclips.net/video/Qe8NQx4ibE8/видео.html
019 Hartmann Locks
Young Lisa Miller Gary Hernandez Thomas