Understanding Decimation: Downsampling Signals

This page describes the decimation process with figures and examples, and also mentions the MATLAB command for decimation.

Decimation is the process that decreases the sampling rate of a signal. Hence, it’s also called Down sampling. This is achieved by removing or dropping samples.

It’s crucial to check for any violation of the Nyquist-Shannon sampling theorem during decimation.

decimation

Decimation Example

Let’s consider an example:

X(n) = [0 1 2 3 4 5 6 7 8], at a sample rate of Fs. If the decimation factor (D) is 2, then the decimated signal Y(n) will be:

Y(n) = [0 2 4 6 8], at a sample rate of Fs/D.

MATLAB Command for Decimation

The resample command in MATLAB can be used for decimation.

>> resample(IQ_25, 4, 25)


Here:

- `IQ_25` is the IQ data at 25 MSPS (Mega Samples Per Second).
- `IQ_4` (the result) is the IQ data at 4 MSPS.

The `resample` function effectively changes the sampling rate of `IQ_25` from 25 MSPS to 4 MSPS.