As others have said, noise is broadband unless it is filtered somewhere. Without any filtering, all of the noise power will alias into the baseband when you perform your sampling and it will raise the noise floor of your measurement / recording.<br><br>In general you don't want the bandwidth of you amplifier to be any greater than it needs to be in order to capture all of the signal, otherwise it will just let through unwanted noise. And the wider you open the gate, the more noise that will flood in.<br><br>A recent commercial project I was working on used strain gauges for some wireless instrumentation. These had gains upwards of x4000, and although the strain signals themselves had a quite controlled and predictable bandwidth, the bandwidth of the amplifiers needed to be tightly controlled to prevent excessive measurement noise.<br><br>-Richie, <br><br>Sent from my Xperia SP on O2<br><br>---- Steve wrote ----<br><br><div style="font-family: Verdana;font-size: 12.0px;"><div>
<div>Howdy @ DSP experts out there,</div>
<div> </div>
<div>I was wondering:<br/>
if done by the book, if you're doing ADC conversions of an audio signal and like it to be clean, you put a nicely steep LPF before it that, immensely reducing anything >= fsample/2.</div>
<div><br/>
So somebody told me recently, he wants to analyse an audio signal which pretty much only has frequencies up to 4kHz, maybe 6 but very quiet, and he's interested only in 1..2 kHz of that.</div>
<div>Hence, he wants to omit the LPF to save parts & PCB space, sample at a rate like 15..25 kHz and use a digital filter and then decimate to 4 kHz or so.</div>
<div> </div>
<div>Is that really feasible?<br/>
The thing is, the signal is picked up by an analog mic (piezo) with 2 opamps after another, each has an integrated PGA of up to 16x gain, because the signal get get really quiet.<br/>
<br/>
Now even if the sound source picked up and the frequency response curve of the piezo make sure that the mentioned upper limit holds.<br/>
Could that stuff not pick up noise in some environments which introduce frequencies above 1/2 of his higher samplerate and hence pollute the spectrum?<br/>
<br/>
Is it *ever* a good idea to omit the analog filter before sampling?</div>
<div> </div>
<div>- Steve<br/>
</div>
</div></div>