From: buttercup on
Hello,

I'm currently attending a class on basics of signal processing.

There's something I don't quite understand:
we model the ideal sampler of a function s(t) as multiplication of an
impulse train with that function.
Why a multiplication? It seems counterintuitive to me, considering a
sampler is supposed to obtain the values of a function at discrete
times.

Wouldn't a convolution be more appropriate?

Thanks in advance for your time!