AndyN wrote: > If I understand correctly; for the Nyquist sampling to result in a > perfectly reconstructible waveform (at the receiving end) the sampled > waveform must be repetitive and the waveform must be sampled for a > period that is long in comparison to the waveform.
The requirement for perfect reconstruction is that the analog signal be -strictly bandlimited- to a frequency less than half of the sample rate. In mathematical terms, "strictly bandlimited" means the magnitude of the Fourier transform of the signal must be identically zero for f >= fs/2. This does not require periodicity. However, such a signal will be infinite in duration because the impulse response of the ideal "brick wall" filter assumed by the sampling theorem (a sinc finction) is infinite in extent. Actual filters that approximate this usually have finite impulse response duration, so this is not a problem in practice, but does cause small errors from perfect reconstruction. ------------------------------------------------------------------------ andy_c's Profile: http://forums.slimdevices.com/member.php?userid=3128 View this thread: http://forums.slimdevices.com/showthread.php?t=95771 _______________________________________________ audiophiles mailing list audiophiles@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/audiophiles
