<< Chapter < Page | Chapter >> Page > |
In this module we illustrate the processes involved in sampling and reconstruction. To see how all these processes work together as a whole, take a look at the system view . In Sampling and reconstruction with Matlab we provide a Matlab script for download. The matlab script shows the process of sampling and reconstruction live .
To sample an analog signal with 3000 Hz as the highest frequency component requires samplingat 6000 Hz or above.
The sampling theorem can also be applied in two dimensions, i.e. for image analysis. A 2D sampling theorem has a simple physical interpretation in image analysis:Choose the sampling interval such that it is less than or equal to half of the smallest interesting detail in the image.
We start off with an analog signal. This can for example be the sound coming from your stereo at home or your friend talking.
The signal is then sampled uniformly. Uniform sampling implies that we sample every ${T}_{s}$ seconds. In we see an analog signal. The analog signal has been sampled at times $t=n{T}_{s}$ .
In signal processing it is often more convenient and easier to workin the frequency domain. So let's look at at the signal in frequency domain, . For illustration purposes we take the frequency content of the signal as a triangle.(If you Fourier transform the signal in you will not get such a nice triangle.) Notice that the signal in is bandlimited. We can see that the signal is bandlimited because $X(i)$ is zero outside the interval $\left[-{}_{g} , {}_{g}\right]$ . Equivalentely we can state that the signal has no angular frequencies above ${}_{g}$ , corresponding to no frequencies above ${F}_{g}=\frac{{}_{g}}{2\pi}$ .Now let's take a look at the sampled signal in the frequency domain. While proving the sampling theorem we found the the spectrum of the sampled signal consists of a sum of shifted versions of the analog spectrum. Mathematically this isdescribed by the following equation:
In we show the result of sampling $x(t)$ according to the sampling theorem . This means that when sampling the signal in / we use ${F}_{s}\ge 2{F}_{g}$ . Observe in that we have the same spectrum as in for $\in \left[{-}_{g} , {}_{g}\right]$ , except for the scaling factor $\frac{1}{{T}_{s}}$ . This is a consequence of the sampling frequency. As mentioned in the proof the spectrum of the sampled signal is periodic with period $2\pi {F}_{s}=\frac{2\pi}{{T}_{s}}$ .
So now we are, according to the sample theorem , able to reconstruct the original signal exactly . How we can do this will be explored further down under reconstruction . But first we will take a look at what happens when we sample too slowly.
If we sample $x(t)$ too slowly, that is ${F}_{s}< 2{F}_{g}$ , we will get overlap between the repeated spectra, see . According to the resulting spectra is the sum of these. This overlap gives rise to the concept of aliasing.
The consequence of aliasing is that we cannot recover the original signal,so aliasing has to be avoided. Sampling too slowly will produce a sequence ${x}_{s}(n)$ that could have orginated from a number of signals. So there is no chance of recovering the original signal.To learn more about aliasing, take a look at this module . (Includes an applet for demonstration!)
To avoid aliasing we have to sample fast enough. But if we can't sample fast enough (possibly due to costs) we can include an Anti-Aliasing filter. This willnot able us to get an exact reconstruction but can still be a good solution.
The stagecoach effect
In older western movies you can observe aliasing on a stagecoach when it starts to roll. At first the spokes appear toturn forward, but as the stagecoach increase its speed the spokes appear to turn backward. This comes from the fact that the sampling rate,here the number of frames per second, is too low. We can view each frame as a sample of an image that is changing continuouslyin time. ( Applet illustrating the stagecoach effect )
Given the signal in we want to recover the original signal, but the question is how?
When there is no overlapping in the spectrum, the spectral component given by $k=0$ (see ),is equal to the spectrum of the analog signal. This offers an oppurtunity to use a simple reconstruction process. Remember what you have learned about filtering.What we want is to change signal in into that of . To achieve this we have to remove all the extra components generated in the sampling process.To remove the extra components we apply an ideal analog low-pass filter as shown in As we see the ideal filter is rectangular in the frequency domain. A rectangle in the frequency domain corresponds to a sinc function in time domain (and vice versa).
Then we have reconstructed the original spectrum, and as we know if two signals are identical in the frequency domain, they are also identical in the time domain . End of reconstruction.
The Shannon sampling theorem requires that the input signal prior to sampling is band-limited to at most half the sampling frequency. Under this conditionthe samples give an exact signal representation. It is truly remarkable that such a broad and useful class signals can be represented that easily!
We also looked into the problem of reconstructing the signals form its samples. Again the simplicity of the principle is striking: linear filtering by an ideal low-pass filter will do the job. However,the ideal filter is impossible to create, but that is another story...
Go to?
Notification Switch
Would you like to follow the 'Fundamentals of signal processing' conversation and receive update notifications?