Signal to Noise Ratios
Posted By Shawn Billings on 8/23/2001 at 1:04 PM

In version 2.4 signal to noise ratios can be evaluated graphically in table form. The table has values on the left side and a UTC or local time line on the bottom. How should we interpret the values on the left side of the table. My brief experience has been that these values (using Locus) have been betwee 35-60. It would also appear that the signal to noise ratios in the 50's seem to process better than those in the 30-40 range. What do these values represent and how do the users benefit from this knowledge?

Thanks
Shawn



Re: Signal to Noise Ratios
Posted By James Webb on 8/23/2001 at 8:21 PM

Shawn,

Put simply enough for me to understand, the S/N ratio effectively means bigger number better reception, smaller number worse reception.

How to use it effectively....I'm not sure but here goes, kinda OTF thinking...

Assume a long static session. If for some reason....ionosphere, flock of 100000 geese circling, A-10 fighters targeting you, or whatever.....the signal strength drops off as shown by a dip in the S/N ratio then you might improve the accuracy by cutting this portion out of the solution.

What is the "magic" number for a cut-off ? I have no idea.

Jimbo



Re: Signal to Noise Ratios
Posted By Greg Shimp on 8/24/2001 at 6:11 PM

I think it's a relative thing, If you have 7 SVs in the 50ish range, and 1 SV bouncing around in the 30-40 range, it's not good, relatively speaking.

I think the more interesting part of the "signal to noise ratio" plot is that it shows "loss of signal" and "possible loss of signal". Perhaps one of the experts can comment on this, but I would assume that frequent loss of signal is more problematic than a low signal to noise ratio. Anyone?

-Greg



Re: Signal to Noise Ratios
Posted By Dave Huff on 8/24/2001 at 8:23 PM

Hey Guys,

I seem to remember from the manual that the thing to look for in the "signal to noise" ratio is not the value, but for "widely differing values".



Re: Signal to Noise Ratios
Posted By J.D. Billings on 8/24/2001 at 9:59 PM

Dave,

Can you quantify "widely differing values"? I think that's what we're looking for.

Thanks

J.D.