Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: noise voltage (was: Conservation Laws)



At 08:51 PM 3/15/00 -0600, brian whatcott wrote:
>
... reminds me of a comment I think John Denker
made some weeks ago - about equipment to dig out signal from the noise.
At that time, I thought he was somewhat overstating the case for the GPS
for example:
it is a given that commercial GPSs can lock functionally with a signal plus
noise at least as big as -170dBm and a low noise input amplifier can hardly
get much better than 3microvolt.Hertz.

I don't know where that 3microvolt.Hertz number is coming from. Certainly
not from me.

I tend to think of noise as being more like
nanovolts per root Hertz (into 50 ohms)

This involves
a) orders of magnitude smaller numbers
b) frequency in the denominator, not numerator, and
c) square root, not first power, of frequency.

The question of what -170dBm is meant to be compared to is another point.
Of course the answer is 1 milliwatt

Right.

So it all depends on how much instantaneous bandwidth that receiver is
using - not a lot, I don't suppose.

Right. If you're willing to work at it for a second or so, (i.e. a 1Hz
bandwidth), you can definitely dig out a -170dBm signal, given a good
receiver front end.

And the typical GPS signal is usually quoted to be -160dBm, which is a
huuuge improvement over -170dBm.