Dear Raghunathan,

decimating is a methode to reduce the number of samples, means the amount of data of a measuring application. Decimating is usually combined with a desired low pass filtering, so that a reduction of bandwidth can be combined with a reduction of sample rate, without loosing information. Remember a temperature measurement, for instance, where the physical situation does not present any fast changes, but where the fast changes in the signal is unwanted noise, which is wished to be filtered out. In this situation there's no need for a high sample rate after the low pass filtering, so that a low pass filtering in combination with a decimating makes sense.

Decimating works with FIR-filters (finite pulse response), where a number of samples in a row are weighted by certain coefficients and added together (integrated), which the moving average filter is a special case of (all samples weighted by factor 1).

Decimating with IIR-filters (inifinte impulse respsonse) on the other hand is much harder. You must calculate all samples. Or concretely:

In the case of FIR-filters, any output is a function only of the past inputs (because there is no feedback). Therefore, you only have to calculate outputs which will be used.

For IIR-filters, you still have to do part or all of the filter calculation for each input, even when the corresponding output won't be used. The reason is that outputs you do use are affected by the feedback from the outputs you don't use.

Downsampling is not the same as decimating, but the process of throwing away samples without low pass filtering. A signal can be downsampled (without doing any filtering) whenever it is "oversampled", that is, when a sampling rate was used that was greater than the Nyquist criteria required.

Kai