4
$\begingroup$

I am new to this site and field too.I always thought analog is continuous and digital signal is discrete. I read this today and got confused about analog signal

Analog can be continuous time(CT) or discrete time(DT). A discrete time signal is not a digital signal; a discrete time signal is one in which the amplitude is a continuum, and not discretized. On the other hand, if the time is discretized, this signal is still analog. Only when a discrete time signal is passed through an A to D converter, it becomes a digital signal.

Will anybody explain me what actually it means?

Improve this question
$\endgroup$
1
  • 1
    $\begingroup$ filters can be made outa Charge-Coupled Devices (CCD) a.k.a. "bucket brigade" and would be an example of discrete-time and analog. $\endgroup$
    – robert bristow-johnson
    Jan 2, 2015 at 19:43

1 Answer 1

Reset to default
7
$\begingroup$

This sounds like a confusion in terminology. See http://en.wikipedia.org/wiki/Digitizing

Digitization basically involves two steps:

  1. Discretization: Sampling the signal at discrete times
  2. Quantization: Turning the samples from (in theory) infinite resolution to finite resolution

Look at all possible combinations you can have 4 different types of signal:

  1. continuous time, continous amplitude
  2. discrete time, continous amplitude
  3. continuous time, discrete amplitude
  4. discrete time, discrete amplitude

Typically we refer to #1 as "analog" and #4 as "digital". #2 and #3 are of little practical relevance, other than getting the mathematical models right. Quantization and Discretization are really quite different mathematically so it doesn't make a lot of sense to analyze them in combination. In other words: the theory of time discretization applies equally to continuous amplitude and discrete amplitude signals.

Improve this answer
$\endgroup$
4
  • $\begingroup$ so whether #2 and #3 are combination of analog and digital signal? $\endgroup$
    – user3857907
    Jan 8, 2015 at 14:37
  • 1
    $\begingroup$ An important case of "almost" #2 in your list is a high precision algorithm implementation, where the quantization is absolutely neglegible (of no consequence to the performance of the algorithm). This is typically a prototype implementation, which may later be transformed into a reduced wordlength implementation to reduce the hardware complexity (make it cheaper). $\endgroup$
    – Jens
    Jan 26, 2015 at 19:54
  • $\begingroup$ Real world #2 is en.wikipedia.org/wiki/Switched_capacitor $\endgroup$
    – endolith
    May 22, 2017 at 20:54
  • $\begingroup$ @user3857907 Take a look at this answer: dsp.stackexchange.com/a/34656/26009 $\endgroup$
    – Envidia
    May 22, 2017 at 22:43

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.