Can anyone quantify for us the values shown in the Signal to Noise Ratio graph in ver 2.4, and how to best interpret them?

J.D.

JD,

After reading this...........

"· Segments of satellite data that contain multiple flags. This is characteristic of an obstructed satellite.

· Gaps in the data caused by extended loss of lock of the satellite. This is characteristic of an obstructed satellite.

· A satellite with rapidly varying signal-to-noise ratio compared to other satellites. This is characteristic of a satellite affected by multipath or an active ionosphere.

· A segment of a satellite with rapidly varying signal-to-noise ratio compared to the rest of the same satellite data. This is characteristic of a segment of a satellite affected by multipath or an active ionosphere.

· A satellite contributing a very small amount of data compared to the other satellites in the data set. Sometimes such a satellite causes problems with processin"

I'm not sure the actual numerical value is important, of course provided it is above a minimum threshhold and is useable. I assume the possible loss of lock and the loss of lock flags provide us with a method of knowing when the actual signal strength is not useable.
( I hate assuming...you know why I'm sure.)

I think the short time contributions, gaps(easily understood) and the rapid changes in signal strength as shown in the signal to noise ratio plot are the keys to using this tool.

What is a rapid change......... ???

Just guessing.....

Loss of lock and possible loss seem to be flagged at about a 10% or greater change in my S/N Plot. This is from one epoch to the next.

I wish somebody would show us a graph of a multipath effect. I see a lot of small rapid changes epoch to epoch (1 or 2 points) on the graph but I don't think this is multipath. all three units do it and it is throughout the session, sometimes "stepping" in one direction or the other through several epochs.(But I think this is normal).

Help..............

Jimbo

I don't know if this is the case with your data but here is something to keep in mind when looking at residuals. You are looking at residuals of double difference quantities. If you see common errors in your residuals for all satellites at a station at the same epoch it doesn't necesarily mean that there is an error in all satellite observations at that station at that time. If there was an error in the reference satellite observation at the station at that epoch it will show up in the residuals of all double different observations for that epoch at that station.

Similarly if you see the same thing happening to a particular satellites observations (not the reference satellite) at all stations at the same epoch it could mean that there was something wrong with the observations of that satellite at the reference station only.

Finally, if you see the same thing happening to all satellites at all stations at the same epoch this could mean that something happened to the reference satellite at the reference station only at that time.

The above sounds complicated but all it is saying is that when you look at double difference residuals for a particular satellite at a particular epoch they reflect what is happening to observations of both satellites at both stations that are used to form the double differences, ie the reference satellite and the reference station as well as the station and satellite that may appear on the plot.

The manual says that the S/N ratio should average 50, whatever 50 is in the context of the plot.

I have found that the S/N ratio plot for a given satellite during the course of its trajectory across my sky forms an arc, with the sat appearing on the horizon at "30" and climbing to "50" or "51" and then falling to "30" or so before setting.

The software appears to choose the highest sat for reference, until it begins to wane.

This makes sense. The thing I wonder is, why does the plot jog up and down with every epoch, as though the receiver only sees it every other time around. The raw data always shows a vertical zig-zag for S/N. What is that?

on page 65 ;>)
if its rapidly varying vs.

a. other satellites
b. same satellite other times

this indicates multipath. So I guess its a very relative/subjective call to make this distinction.

Jimbo
EDIT: I never emphasized this before in my mind when reading it the first round.
Modified By James Webb on 8/29/2001 at 9:58 PM

Maybe I can help to conceptualize the ups and downs in the signal to noise plots.

The two components coming into the radio receiving section of the GPS receiver are signal and random noise. The signal strength is affected by obstructions, which cause outages or questionable data flags. The effect of mutipath and/or ionosphere in the signal to noise values and residuals are usually 5-15 minute periods of sinusoidal variation. Real pronounced examples of multi-path and ionosphere are obvious in the residual plots and can affect the receivers' ability to resolve the centimeter level solution.

Background RF noise in the GPS frequency spectrum causes the epoch by epoch jumps in the signal to noise ratio. Keep in mind that the power of the GPS signal is far less than the random background noise emitted by the Earth. The GPS signal power as it is transmitted is about the same as a 40 watt light bulb.

What a great system!

Cheers,
Rod Johnson
Ashtech Marketing

Rod,

It does help. Knowing the effects of multipath would tend to be like a sine wave gives us "newbies" a good clue as to what we are looking at in the S/N plot.

I had kinda guessed the epoch by peoch jumps would tend to be noise related. But I am surprised the SV produces that strong a signal. I thought they were closer to 4 watts, really weak.

Thanks, keep watching, keep posting. I need all the help I can get.

Jimbo