# Subtract audio from noise

A

#### Ancient_Hacker

Jan 1, 1970
0
Why do you you think a simple detect/invert/sum analog process needs "very
regular and predictably structured noise?"

The ones I've used work well with random noise.

If you add two random noises, you just get more random noise.

By definition, random isnt correlated with anything else, especially
other random sequences.

Without exact algebraic correlation, you can't subtract and get a
reduction.

Now you can clip, divide, or blank impulse noise, but that's not noise
in the hissy sense. It's reliably short and spiky and above the
desired signal level, making it reliably detectable. You can't do
the same for hissy noise, by definition. If you could, then we'd be
able to pick up several million TV stations from other galaxies with
just a tea saucer covered with tn foil.

D

#### Don Bowey

Jan 1, 1970
0
If you add two random noises, you just get more random noise.

And if both random noises are from the same source at the same time?
By definition, random isnt correlated with anything else, especially
other random sequences.

Certainly it correlates to itself......
Without exact algebraic correlation, you can't subtract and get a
reduction.

Now you can clip, divide, or blank impulse noise, but that's not noise
in the hissy sense. It's reliably short and spiky and above the
desired signal level, making it reliably detectable. You can't do
the same for hissy noise, by definition. If you could, then we'd be
able to pick up several million TV stations from other galaxies with
just a tea saucer covered with tn foil.

Atmospheric noise, in which I include galactic white noise, covers a wide
spectrum, and what is detected at frequency A will also be detected at
frequency B.

If I am listening to a radio station at frequency A (with atmospheric
noise), and my "noise blanker" is tuned to a clear frequency (B), receiving
the same atmospheric noise, it will work fine.

A

#### Ancient_Hacker

Jan 1, 1970
0
And if both random noises are from the same source at the same time?

Certainly it correlates to itself......

Atmospheric noise, in which I include galactic white noise, covers a wide
spectrum, and what is detected at frequency A will also be detected at
frequency B.

????? You're telling me that a receiver tuned to 1000 MHz with a
bandwidth of 1KHz is going to hear noise. So far so good.
But what makes you think that noise is in any way correlated with
noise at any other frequency or bandwidth?

The spike at 1.2345 volts at 1000MHz may have come from an atom of
hydrogen at alpha centauri jumping from level Q to F.

The spike of 2.77 volts at 1001MHz may have come from an atom of
iron from the crab nebula 1054 light years away in a different
direction from both us and the aforementioned atom.

The spike of 4.223 volts at 1002MHz may have come from the spark plug
of the Harley down the street.

There's no way they're all in any way shape or form related.

And even
----

Now I know it "seems" to be that way but that's only because Harleys
put out broadband noise that IS often correlated.
Even so, there's not a bit of "subtraction" going on-- it's a noise
BLANKER, which just punches a hole in the noise-- no way it subtracts
out the exact amount of noise and leaves the signal.

D

#### Don Bowey

Jan 1, 1970
0
(snip because the ancient one made a mess of it)
????? You're telling me that a receiver tuned to 1000 MHz with a
bandwidth of 1KHz is going to hear noise. So far so good.
But what makes you think that noise is in any way correlated with
noise at any other frequency or bandwidth?

!!!!!!!! No, I went back to look and I definitely did NOT say that. I had
around 40 MHz. in mind for the noise channel.

I didn't read the rest of your post, as it is clear you are too devious for
me.

(snip)

Replies
4
Views
596
O
Replies
12
Views
4K
Jasen
J
F
Replies
5
Views
1K
Ken Smith
K
F
Replies
18
Views
7K
Don Bowey
D
Replies
9
Views
112