what is the " alliasing" and "folding " of a signal and how can be overcomed while using filters?

Answer 1

In order to correctly identify the frequency of detect an oscillating signal, your detector must be able to sample at least twice per oscillation. This condition (the Nyquist theorem) places an upper limit on the frequencies that can be distinguished by your detector. If your signal contains oscillations which are faster than this limit, your detector will erroneously identify the signal as having a lower frequency (aliasing). Aliasing can be avoided by introducing filters which remove signals whose frequency is above the limit of the detector.

Answer 2

The most graphic demonstration of "aliasing" occurs in motion pictures containing a moving wheel, which seems to be rotating backwards. The biggest hazard in "aliasing" is that some engineers apply filters without analyzing what they are actually doing. If they intent to re-construct sound, they are probably right. If they are collecting time spaced measurements for instrumentation, they are probably off base.