0
$\begingroup$

The general structure of Sub-Band Coding (SBC) is shown below: enter image description here

The diagram shows that after each of the analysis filters, decimation is performed by a factor M which corresponds to the number of channels. Would this not result in aliasing when this decimation by N is performed for the higher frequency sub-bands and why is DDC not required in each path to translate each sub-band to baseband before performing the decimation and thus avoiding aliasing?

$\endgroup$
0
1
$\begingroup$

I think you're just misjudging what the [CHANNEL] block does: this is just saying "all the bits that fall out of the different quantizers/encoders get transmitted".

It's especially not an adder! Most books actually show this just as a "frame packer" or "interleaver".

$\endgroup$
4
  • $\begingroup$ Thank you for your response. My ambiguity is around the portion where decimation takes place, would that decimation not result in aliasing as the Fs would be reduced for each path? $\endgroup$ – malik12 Feb 18 at 11:28
  • $\begingroup$ that's a different question than what you're asking above, but what do you think is the job of the analysis filters? $\endgroup$ – Marcus Müller Feb 18 at 11:32
  • $\begingroup$ I am sorry if I was not able to explain my question correctly. My idea is that the analysis filters separate the frequency bands but in that filtering the sample rate is still same as original. For example say my Fs is 100 MHz and I use analysis filters at 20-40MHz and 60-80 MHz. After I decimate the filtered signal by say a factor of 2, would not that cause aliasing for these bands? $\endgroup$ – malik12 Feb 18 at 11:43
  • $\begingroup$ you need to think about the relationship between your decimation and the bandwidths of these filters. $\endgroup$ – Marcus Müller Feb 18 at 11:46

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.