ASN7 - Sampled half of signal: Difference between revisions
Jodi.Hodge (talk | contribs) No edit summary |
Jodi.Hodge (talk | contribs) No edit summary |
||
(5 intermediate revisions by the same user not shown) | |||
Line 1: | Line 1: | ||
[[Jodi Hodge|Back to home page]] |
[[Jodi Hodge|Back to home page]] |
||
⚫ | |||
Can you retrieve an original signal from a sample of range <math> \frac{1}{2}f>f_s>f \!</math>? |
|||
Lets emagine that from our sampling we splitted the positive and the negitave part of the signal. So now we have the negitive half repeated on the negitive axis and the positive half all along the positive axis. |
|||
⚫ | |||
⚫ | |||
⚫ | |||
'''Follow this procedure''' |
|||
1. Lets imagine that from our the result of our sampling we measured the positive and the negative part of the signal seperately. [Assume the original signal was centered at zero such that <math> f_{tot}= f_{-}+f_{+} \!</math> |
|||
So we have repeated on the negative axis the -f and the positive axis +f. |
|||
⚫ | |||
2. Find the right frequency and sample again. |
|||
The result of this second sampling are a number of copies of the original signal <math>nf_{tot} \!</math> . |
|||
⚫ |
Latest revision as of 12:36, 20 December 2009
Can you retrieve an original signal from a sample of range ?
A rule of thumb for proper sampling is to sample range slightly greater than the range of the signal. Suppose however,you sampled only half of the signal can you do something to recover the whole original signal again?
Follow this procedure
1. Lets imagine that from our the result of our sampling we measured the positive and the negative part of the signal seperately. [Assume the original signal was centered at zero such that
So we have repeated on the negative axis the -f and the positive axis +f.
The trick now is to re-sample at a frequency that will place the negative and positive half side by side.
2. Find the right frequency and sample again.
The result of this second sampling are a number of copies of the original signal .
3. Finally, to get just one copy of the original signal apply a band pass filter.