I work tangentially with wireless communications signals in my day job and although I have a good innate understanding of FDMA and TDMA, I was having a hard time grasping the concept of CDMA. This video gave me the clear explanation I was looking for, thank you!
Now the challenge of synchronizing between the symbol signals and the sequence codes is real challenge. Also what makes it more challenging is having long sequence and knowing when to switch to the next sequence code, syncing is vital or else we lose the received data. It would be greatly appreciated if you could explain how to perform this syncing between code sequence and the correlated original symbol signal data.
@@iain_explains I understand your very busy schedule ... Thank you for all of your great efforts you have done so far. I am one of your fans and am always watching your great videos.
Hello are we mixing the spread sequence with the bits together? I am wondering how this mixing would result in the frequency domain... I have read from books (what you have also mentioned in the video) that the the resultant spectrum is wider and makes the signal more resistant to interference. But how does that work mathematically I am wondering. Last but not least, this is a great video 👍
I'm not sure what you're asking, exactly. Multiplication in the time domain is the same as convolution in the frequency domain. That's the mathematical relationship. Perhaps these videos will help: "Convolution in the Time Domain is Equivalent to Multiplication in the Frequency Domain" ruclips.net/video/Ex318cCXSBM/видео.html and "Amplitude Modulation AM Radio Signal Transmission Explained" ruclips.net/video/-PWg-0k2oks/видео.html
Hi Sir! Great video and thanks for uploading! Just for my clarification, the psd of the spread signal would be for one chip not for the entire 6 chips, correct?
I think what you're trying to ask is whether the wide frequency width of the PSD is a result of the short time period of the chip waveform. If so, then the answer is Yes.
A 6-chip waveform is just a 1-chip waveform added to 5 other 1-chip waveforms each offset by an increasing number of chip intervals. Time offsets are just phase changes in the frequency domain, and so the bandwidth of the 6-chip waveform is the same as the bandwidth of the 1-chip waveform.
Thanks for the suggestion, I'll put it on my "to do" list. In summary, each delayed path in the channel (ie. the ISI paths) will arrive at the receiver with a different time offset, and can be correlated with the spreading code at that respective time offset, in order to detect the data on each path separately. If the spreading code is orthogonal to time-shifted versions of itself, then they won't interfere with each other.
This is wonderful. Thanks for uploading this on my previous request. Had few questions in mind, I was referring "Wireless Communication" book by Rappaport and found out that we need to use the equalizer at Receiver for the time varying channel. How those would affect in this DS-CDMA case ? And i think RAKE receiver would be a way to go as receiver for most of the CDMA system. Is there any simple way for designing a receiver ? Simply let's say we have ADALM PlutoSDR, can we use that as receiver, which i believe is not RAKE receiver. If you can comment on these questions, would be helpful. Thanks for your great work.
Sorry, I'm not familiar with the particular SDR you mentioned. In general though, yes, you're right, typically DS-CDMA is received with a rake receiver. I've got it on my "to do" list.
Would this be able to work with chirps? So for instance using Generalized-Chirp like (GCL) Polyphase codes/sequences. In order to closely resemble LoRa..
Sorry, I probably didn't make it clear that I'm assuming a scaling here. Technically, we should be writing down the value of the output of a matched filter in each case. So, the non-spread pulse would be received as (+1)(+1)(T) = T. Each chip in the spread pulse would be received as (+1)(+/-1)(T/6) = +/-(T/6), and then they get summed up. So it's all OK in the video, just called by T.
I work tangentially with wireless communications signals in my day job and although I have a good innate understanding of FDMA and TDMA, I was having a hard time grasping the concept of CDMA. This video gave me the clear explanation I was looking for, thank you!
That's great to hear. I'm so glad it was helpful.
Now the challenge of synchronizing between the symbol signals and the sequence codes is real challenge. Also what makes it more challenging is having long sequence and knowing when to switch to the next sequence code, syncing is vital or else we lose the received data. It would be greatly appreciated if you could explain how to perform this syncing between code sequence and the correlated original symbol signal data.
Thanks for the suggestion. It's on my "to do" list (but it's a long list).
@@iain_explains I understand your very busy schedule ... Thank you for all of your great efforts you have done so far. I am one of your fans and am always watching your great videos.
Really grateful for your videos, you do a great job!
Glad you like them!
Hello are we mixing the spread sequence with the bits together? I am wondering how this mixing would result in the frequency domain... I have read from books (what you have also mentioned in the video) that the the resultant spectrum is wider and makes the signal more resistant to interference. But how does that work mathematically I am wondering.
Last but not least, this is a great video 👍
I'm not sure what you're asking, exactly. Multiplication in the time domain is the same as convolution in the frequency domain. That's the mathematical relationship. Perhaps these videos will help: "Convolution in the Time Domain is Equivalent to Multiplication in the Frequency Domain" ruclips.net/video/Ex318cCXSBM/видео.html and "Amplitude Modulation AM Radio Signal Transmission Explained" ruclips.net/video/-PWg-0k2oks/видео.html
@@iain_explains "Multiplication in the time domain is the same as convolution in the frequency domain." yea that's what I mean exactly.
Hi Sir! Great video and thanks for uploading! Just for my clarification, the psd of the spread signal would be for one chip not for the entire 6 chips, correct?
I think what you're trying to ask is whether the wide frequency width of the PSD is a result of the short time period of the chip waveform. If so, then the answer is Yes.
@@iain_explains thanks Sir! For further clarification. When u say 1 chip waveform do u mean 1 chip and not 6 chip, correct?
A 6-chip waveform is just a 1-chip waveform added to 5 other 1-chip waveforms each offset by an increasing number of chip intervals. Time offsets are just phase changes in the frequency domain, and so the bandwidth of the 6-chip waveform is the same as the bandwidth of the 1-chip waveform.
Thanks for this wonderful video. Could you further explain how DSSS makes Inter-Symbol Interference negligible in the multi-path scenario?
Thanks for the suggestion, I'll put it on my "to do" list. In summary, each delayed path in the channel (ie. the ISI paths) will arrive at the receiver with a different time offset, and can be correlated with the spreading code at that respective time offset, in order to detect the data on each path separately. If the spreading code is orthogonal to time-shifted versions of itself, then they won't interfere with each other.
This is how teaching should be!
I'm glad you liked it.
This is wonderful. Thanks for uploading this on my previous request. Had few questions in mind, I was referring "Wireless Communication" book by Rappaport and found out that we need to use the equalizer at Receiver for the time varying channel. How those would affect in this DS-CDMA case ? And i think RAKE receiver would be a way to go as receiver for most of the CDMA system. Is there any simple way for designing a receiver ? Simply let's say we have ADALM PlutoSDR, can we use that as receiver, which i believe is not RAKE receiver. If you can comment on these questions, would be helpful. Thanks for your great work.
Sorry, I'm not familiar with the particular SDR you mentioned. In general though, yes, you're right, typically DS-CDMA is received with a rake receiver. I've got it on my "to do" list.
Would this be able to work with chirps? So for instance using Generalized-Chirp like (GCL) Polyphase codes/sequences. In order to closely resemble LoRa..
It'll work for any waveform that spreads the bandwidth and is orthogonal to the waveforms being employed by other users.
very well explained, thank u !
Glad it was helpful!
CDMA seems to be a great idea (theoretically) but what make people abandoned CDMA and move on to OFDMA ?
Did you see my video comparing them? "Mobile Standards Evolution: FDMA, TDMA, CDMA, OFDMA" ruclips.net/video/bm53RpK-S2k/видео.html
@@iain_explains thank you so much !
Thanks! 😁
You're welcome!
I never understood CDMA in its core principle, not I got a clear idea thanks
Glad it was helpful.
Oh man ... I am now feeling I am 30 years younger LOL
I didn't expect my videos would help to reverse the ageing process, but I'm glad they are 😀
3 pluses + 3 minuses = 0 (I understand this)
6 pluses = 1 (What? How? Shouldn't that equal 6 instead?)
Sorry, I probably didn't make it clear that I'm assuming a scaling here. Technically, we should be writing down the value of the output of a matched filter in each case. So, the non-spread pulse would be received as (+1)(+1)(T) = T. Each chip in the spread pulse would be received as (+1)(+/-1)(T/6) = +/-(T/6), and then they get summed up. So it's all OK in the video, just called by T.