Re: Question For Craig Markwardt re Pioneer 10 Data

"George Dishman" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
>
> "ralph sansbury" <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]
> > George,
> >     I think I have a pretty good idea by now of the
techniques
> > used and their reliability  in detecting the actual
> > doppler changes in radiation from Pioneer 10.
> >    I am curious about the Watola paper that Craig Markwardt
refers to and
> > the summary statement there that the signal is 100 times
stronger
> > than the noise.Have you read this or do you know what is
meant
> > here?
>
> I haven't seen the paper and wasn't reading the thread
> when it was mentioned, I have only seen mentions of it.
>
> If someone says "the signal is 100 times stronger than
> the noise" in the radio business, I would assume he has
> seen a signal to noise ratio of 20dB. This is the ratio
> of the rms power in the signal to the rms power of the
> noise in the passband. A factor of 100 in power would
> be a factor of 10 in voltage (P = V^2 / R) for sine
> waves but the noise is random of course.
>
> The signal level is measured from the AGC signal as
> stated in a number of places on the home page.
>
>
http://spaceprojects.arc.nasa.gov/Space_Projects/pioneer/PNStat.h
tml
>
> We talked about that some months ago. If you know the
> relation between gain and control voltage for a
> variable-gain amplifier and the output level of the
> amplifier, then you can find the input signal level by
> measuring the control voltage.
>
> I think the SNR is mesured by the PLL but I am not sure.
>
> >    This is really the important consideration finally. I am
> > supposing that either the  difference between the observed
> >  sequence of voltages and the fitted sine curve of specific
frequency
> > phase and rms amplitude or succession of these or the
difference
> > between a sequence of fitted 1s and 0s and observed 1s and 0s
etc
> > is the basis for this estimated error.
>
> Bit error rate, the fraction of data bits that get
> changed as a result of the noise, is measured by the
> error correction scheme. See below.
>
> >   eg at successive times eg each nanosecond a 1 in the fitted
> > sequence is paired with a 1 in the observed and similarly for
the 0s but
> > in one out of 100 times there is a mismatch.
>
> Data rates are of the order of tens or hundreds of bits
> per second at these ranges. Are you confusing the data
> with the 256MHz digitised voltage samples?
>
> This is from the Pioneer home page, the contact in April 2001:
>
>
http://spaceprojects.arc.nasa.gov/Space_Projects/pioneer/PNimgs/p
10pic42801_main.jpeg
>
> In the top left you can see
>
>   SNR        +0.812
>
>   AGC        -177.62
>
>   Bit Rate   16
>
> The first two will be dB and the latter is 16 bits per second.
>
> About half way down the right hand side there is:
>
>   Bit error   0.0417
>
>   Nbr errors  16
>
> which means there were 16 bits corrupt in 384 bits received,
> probably the last data block, and that would have taken 24
> seconds to arrive at 16bps.
>
> George
>
>