checking the exact distance

GPS actually works by triangulation of distances measured by the travel time of radio signals (not doppler shift).

See .

Accurate laser rangefinders with 1000+ meter range are readily available for under $300.

I thought you had mentioned recently that WiFi often doesn't fly in a straight line?

You spent hours trying to get a GPS fix under these conditions, and you want to use that as an example of how poor GPS accuracy is?

The markers don't seem to get moved here. A GPS is an excellent way to find a marker that you believe has been put in its correct place. A consumer grade GPS is not the tool for challenging the location of the marker. A professional surveyor, who might be using GPS, is different.

I wouldn't use a GPS to pinpoint my buried treasure, but I can find my car at the trailhead parking lot if I come back to it a different way.

That is covered in some basic GPS literature, and since most printed USGS maps are NAD27, it's handy to know the difference. The online an common computer based tools seem to have settled on WGS84. Since I can print my own maps, I don't use NAD27 at all any more.

Then you need a different tool, don't you? Do you actually get a two foot coverage using online topo maps? I can't get that close.

The GPS is still very handy for correlating the one that you have marked visually and with GPS against the higher precision photo/map that you have located elsewhere.

A combination of my non-mapping GPS and a Google map on my cellphone was quite handy this week. I had no idea what zip code or town I was in. Required granularity? A few hundred feet, maybe larger.

I don't think you want your own DGPS. John Deere Starfire RTK claims sub-inch accuracy. But they have an optimal range of six miles.

You might want a different receiver that allows RINEX post processing.

Post-processing is probably what you want. A little Rinex data captured on site, and at some known point, and then a little computer hash. There are CORS stations not too far from you. I think Pigeon Point is probably closest, although it's hard to say.

But, what are you doing? Are you predicting coverage from a known tower, or selecting the location of a new tower? For a known tower, I would think that the operator would have precise coordinates and heights of available positions. For a new tower, a topo map might be better than fussing with GPS accuracy.

If the GPS accuracy is a a hobby project, CORS post processing would probably be the way to go.

Ummmm, last time I checked, light, wireless, wi-fi, and such all travel in staight lines (unless acted upon by a very large massive gravitational field). What I probably meant was that wi-fi bounces around a lot and that propogation is not always by the direct path.

For accurately locating towers, I use differential GPS from a transmitter on Mt Toro. It's not the best because it doesn't cover inland very far, but is good enough for coastal location. If I can't get a differential fix, I interpolate directly from the STRM data:

have the current corrected 1 arc second data (30 meter) but most of the work I did in the past was with the uncorrected data, which was full of anomalies. Still, it was good enough to verify difficult locations on sides of hills. The propogation software I use: an "elevation grid" and crosshairs feature that's useful for locating map features and topography.

The major problem with using GPS to locate towers is that I often cannot position myself directly under the tower. Even if I could get under the tower, a hand-held GPS under all that metal is not a great idea. The few times when I tried it resulted in wildly inaccurate data points due to reflections. If the cell site is on the roof of a building, I often cannot get access to the roof. So, I have to accurately locate something nearby, measure the distance and direction to the tower, and obtain the real location by calculation.

The good news is that I don't have to do this every often. Locating a tower for propogation studies on flat terrain does not require much accuracy. The location could be off perhaps 100ft in any direction and I would still get a decent coverage plot. It's only the ones that are on the tops or sides of hills that require careful positioning. Also, many radio site owners have taken the trouble to accurate locate their towers for customers, which is a big help.

One of my friends used to repair the competitions units for AutoFarm.

box in the tractor has 3 differential GPS receivers inside. It's run by a PC-104 computer. There's a 900MHz data link from the reference base station, which is really an SC104 DGPS transmitter. The

6 mile limitation is because of the 900MHz link. There are some systems that have used commerical VHF FM links, which will go even farther. It's a fairly complex system.

However, that's not why I want my own differential xmitter system. We use GPS for locating people during events and for emergency services. For example, we had APRS trackers following various officials during the last major election. Normal GPS accuracy is sufficient, but there are situations where I could use some additional accuracy. For example, tracking an individual in a crowd at air shows and track events will allow us to send a runner directly to the person instead of fumbling around the crowd. I've also tried blind flying an RC helicopter that used conventional uncorrected GPS and suspect that an improvement in accuracy near the ground would have been useful.

Yep. That's roughly how GPS cell phones work. They send the GPS delays data to the cell site, which sends the data to a service provider, that crunches the numbers, and returns the lat-long location. Not all cell phone GPS receivers do RINEX or BINEX. |

are also pre and post processors for proprietary data formats to RINEX. | | haven't tried these but it's on my things to do list.

Well, for locating the towers accurately, data correction is probably the right way. Unforunately, I don't have a suitable GPS (but can get one). However, the reason I want my own DGPS transmitter has nothing to do with tower locations.

It's about 95% hobby and 5% for pay at this time. The paying part is growing. |

| | Looks like my pile of Allstar 12 GPS receivers will belch C/A code and phase data suitable for conversion to RENIX. | (from 1999) Sigh. Yet another project.

Yes. Exactly as in ray tracing. There are actually modelling programs out there for simulating a reflective environment. It's commonly used in optimizing avionics antenna location. Unless affected by something, RF travels in straight lines.

As always, there are exceptions. Edge diffraction, as in Fresnel Zones, will cause the beam to appear to "bend". It doesn't actually turn into a curved line, but simply changes direction slightly at the edge point. Before and after the edge point, RF still travels in straight lines.

As with diffractions, one can build an RF lens that will effectively "bend" the RF beam by virtue of the different speeds at which RF travels through different materials. Boundary diffraction, as in RF bouncing off or bending when hitting an atmospheric inversion layer is an example.

There's also Faraday rotation, where the polarization of an RF signal is changed after passing through a strong magnetic field. This effect is commonly used in circulators and isolators. While Faraday rotation doesn't exactly "bend" the signal, I guess one could consider a change in polarization as a major change.

We never did hear back from JB about why he cares more about distance than path loss... Maybe it's just a marketting thing...

That would be a series of straight lines, then, for WiFi? That would be a difficult way to measure what most people perceive as distance.

Okay, I'll stop now. You just don't get it.