Reported SNR of GPS Receivers

My point is that packets with errors should never be used to calculate position, and thus any variance in position cannot be used to determine the error rate. At least I see no reason to assume that digital errors would cause the position to vary.

Also, I should point out that error detection is much easier than error correction, and I assume that GPS has bullet-proof error detection. You are correct that error correction has its limits, but error detection should not be difficult at all.

Brian Willoughby Sound Consulting

GPS

lity of

on

=A0

en

I'm just speculating. I don't know error correction and I don't know the specifics of GPS, so I might be totally lost. I am only projecting the principles I know from other fields, where internal variables can be used to estimate the quality of the desired computed result.

Run

| James E.Thompson, CTO | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona 85048 Skype: Contacts Only | |

Hi Jim, You're in Phoenix, huh?

Not wanting to be accused of not adding something useful to this thread, I'll add the following.

One of the scariest movies ever made was Stanley Kubrick's 'The Shining', starring Jack Nicholson as the main character Jack Torrence. My favorite scene in the movie was the super-weird conversation between Jack and the ghost of Delbert Grady in the Overlook Hotel's men's room. The interior of men's room was painted with an unbelievably garish all-red color. A truly strange, other-worldly, room.

That men's room was modeled after a Frank Lloyd Wright-designed bathroom in Phoenix?s famed Arizona Biltmore Hotel.

The next time I'm in Phoenix I'm gonna visit the Biltmore Hotel to check out that men's room. (I hope Mr. Grady doesn't show up though.)

The idiot greegor can f*ck off and die.

But in many digital systems the relationship between SNR and error rate is nonlinear. There's a knee in the curve, beyond which the error rate rises dramatically. Watching the SNR as it approaches this knee might be useful.

On Feb 3, 9:21=A0pm, TheGlimmerMan Archie > I am currently working on our Archie > battlefield tracking and comms solutions.

G > What kind of security and handlers does G > your employer provide to keep their pet G > sociopath from becoming a public embarassment?

Archie > The idiot greegor can f*ck off and die.

Is it a big locked gorilla cage or a pedded cell?

Archie > I am currently working on our Archie > battlefield tracking and comms solutions.

G > What kind of security and handlers does G > your employer provide to keep their pet G > sociopath from becoming a public embarassment?

Archie > The idiot greegor can f*ck off and die.

G > Is it a big locked gorilla cage or a pedded cell?

Does DRS Technologies or other Defense Contractors hire brain disorder or mental case people for defense work?

You do BRAT? ( What does BRAT stand for? Beyond Line-Of-Sight Reporting and Tracking (US Army) )

It's more like you ARE a BRAT, Archie!

Then again, if you Google battlefield tracking, lots of the hits are for computer games by Electronic Arts and others. Is Archie living a rich fantasy life?

it'd

of

only=20

GPS uses direct sequence spread spectrum (DSSS) which admits even less than QAM and trellis coded (with a demodulation constellation) systems of normal SNR estimation techniques. Also, like Paul Hovanian said the = error rate is very non-linear with SNR and has a distinct fall off the cliff effect (like terrestrial and satellite digital TV). The different chipping patterns are used to separate and distinguish the different satellites (all visible or above the horizon satellites can all be received at the same time on a single antenna). Also, packets ala Ethernet do not exist. Unique Words ala LAPB are at a higher decoding level than the base DSSS demodulation. SNR does not really make sense = for GPS signals, even error rate does not apply cleanly.