View Single Post
  #180 (permalink)  
Old 07-07-2009, 10:34 PM
Doc007's Avatar
Doc007
Lurker
Offline
 
Join Date: Apr 2009
Posts: 3
Reputation: 0
Doc007 is a n00b
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Re: (Possibly) Working aGPS Solution! (Take Two)

Great software idea!! =D> I have yet to try it but I can see how this would work. This reminds me of an old convo I had with a fellow member a while back. Perhaps you might be able to implement some of my ideas into your software to help people get a quick lock in a new location that hasn't already been cached. (Old convo is posted below for your convenience. I know its long but I hope you will find it interesting as I did.)

Quote:
Originally Posted by sofladavid View Post
Also. a tip for those having trouble with GPS.

The way GPS works, it that the receiver has to get a "fix" on where you are by receiving signals from at least three satellites. The sattelites are scattered through the sky all around the earth. Obstructions such as Buildings, trees, Roofs, etc will block the signal from being received by the GPS receiver Think about how weak these signals are that are traveling all the way from a satellite in outer space.

When you first turn the GPS on, it hunts around trying to find at least ONE satellite. At that point, the unit has no idea where you are, or what time it is, so it has no idea what sattelite it should be looking for , or where. It's purely trial and error, until it comes across one sattelite that it can communicate with. Once it finds one, it has to download what is called an "almanac". The almanac tells the GPS receiver which other sattelites are in the area and should therefore be listening for. The process of downloading this almanac takes about a minute. If you turn a corner, go under a tree, or otherwise interrupt the downloading of the almanac, the process has to start all over again. luckly, while the almanac is downloading, the GPS is still looking for other satellites and if it finds another one, it starts downloading the almanac from that satellite also. Once it has succesfully downloaded the almanac, then it uses that info to find the other sattelites in the area and determine your location based on the signals it is receiving from the different satellites based on the almanac. The almanac is kept up to date while the GPS is on and connected to the satellites.

If you turn the unit off and then back on a short time later, and you are still in the same general location, then the unit remembers the almanac and knows which sattelites to listen to so it can calculate your position very quickly. But, if it has been a while sine the unit was turned off, or you have moved substantially since the last signals were received, then the almanac is outdated and the unit has to download the alamanac again from scratch (longer time to getting a GPS fix)

This is a very leymans explanation, and I'm sure there may be a few discrepancies (or oversimplifications), but it may help some to understand why sometimes it takes a while to get a GPS fix.

Also, Googlemaps does not give alot of feedback as to what is going on in the background. all you see on the screen is "seeking....."

Most of the handheld GPS units (the GPS apps I am testing too) have a "satellite screen" which shows you what satellites the unit can see, and if they are in "acquisition mode" (downloading the almanac)

Pardon the typos. This is too long to proofread
Quote:
Originally Posted by Doc007 View Post
Thanks for the interesting read on GPS fixing. Maybe you can help me answer my questions based on your tutorial...

Why not have a program calculate the location of the satellites based on user input of a cross street (if you are on a road) or a time + city/town (if you are hiking)?

I figure since these satellites orbit the earth at calculated speed along with the earths rotation. Why not have a program which can make a temporary almanac/offline almanac until a true/real almanac can be downloaded from the primary and/or secondary satellites? Wouldn't something like this reduce the amount of trial and error? Which would result in a reduced acquisition time?
Quote:
Originally Posted by tgreenstein View Post
You're right. Actually, some routing software and GPSr allow you to input a "known" latitude and longitude (or sometimes a city name) from which it starts the process of searching for satellites based on what's called an ephemeris (a fancy name for a table that lists the projected locations of all of the GPS). My Garmin Nuvi 660 does this and some of the standalone PPC software does as well.

Back in the "early" days of consumer GPSr use this was fairly common as the "time to acquisition" of these devices was pretty slow -- 10-20 minutes wasn't uncommon. Current devices can usually obtain a fix in far less time and, probably as a result, don't offer the option to input a "known" location to serve as a starting point.
Quote:
Originally Posted by Doc007 View Post
Thanks for your response.

Would you happen to know which older and/or current PPC softwares utilize ephemeris? I would be interested in trying something like this out in my area where it can take up to 5-6 minutes to aquire a fix from a cold start. The aquisition time is significantly increased while moving. (Likely due to downloading partial almanacs)

I would be interested in seeing if there is a significant difference in aquisition time by using ephemeris in my neighborhood.