Are you measuring inches, feet, or miles? Different methods are used for each. Inches are best done with a pocket tape measure. Feet can be done with either a bigger measuring tape, or perhaps a binocular distance measuring device. Miles are best done with a map.
As far as I know, there's no easy way to use the latency in the hardware to measure distance accurately. I worked on a project that did something like that in an attempt to measure how far away the client radio was from the access point. The idea was to limit users of a wireless hot spot to only those users located on premisis and to block any of the nearby apartment users who just wanted a free broadband connection. It worked, but was deemed to expensive and too unreliable to be useable.
Incidentally, wireless doesn't really go through walls that well. It goes through cracks in the walls and bounces around quite a bit. Using the RF path delay to measure distance in such a reflective environment is not a great idea.
What is the best way to check the exact distance between a laptop and a router? I have tried using a GPS device but it is not very accurate. I assume there isn't a software program that does this -- since it's very hardware related. Is there some other device that could tell me -- it would be through walls and even down the street to find out.
Well, if you had a wire, a TDR or other network tester would tell you, but in the wireless world you don't actually care what the _distance_ is, you care what the path loss is, plus any noise/interference.
Those are a little bit expensive. Much cheaper are little viewers for judging the distance to a golf pin. It relies on having a known height pole at the point being measured, but it might work for you.
Well, the cheap ones are at the golf shop. It's just a scale that measures the height of the "pin" that holds the flag in the hole.
Parallax distance measuring binoculars are useless beyond about
1000ft. If you can increase the baseline to something wider, then you can use these to measure farther. WWI battleships had a 12ft baseline optical range finder. If you have lots of money, a laser range finder works really nice. If you only want to go a few feet, ultrasonic digital measuring tapes are cheap.
It really depends on the accuracy desired. When I do a coverage map using Radio-Mobile software:
formatting link
'm often tempted to use my hand held GPS for locating the site. That never works because GPS alone just isn't accurate enough. When a tower is located on top of a slope (all too common), a slight error in GPS location and the calcs end up using a location that's on the side of the hill instead of the top. Usually, what I end up doing is going to a web site with topo maps:
formatting link
extracting the location from the map instead of from the GPS. This also avoids the problem of different datums. There's quite a bit on handheld GPS location accuracy under the various Geocaching web sites:
formatting link
formatting link
can really see the problem on the Confluence web site:
formatting link
users post pictures of the integer values of latitude and longitude. We have a local confluence near the edge of a local golf course. My guess is that the various photographs over the years cover about an 800ft error radius for finding the exact location. Yech.
When I really need position accuracy for GPS, I use a data logger and software that averages the lat-long over about an hour. I use Visual GPS XP:
formatting link
(free)
formatting link
($25) I can usually interpolate the location to within about 1ft radius after about an hour.
Anyway, if the OP is measuring path lengths in miles, then a few feet inaccuracy isn't going to make much difference in the calculated path length. However, if it's a much smaller distance, then accuracy rapidly becomes a problem.
While I'm ranting on the subject, I've tried WAAS GPS receivers and found some improvement in accuracy. I'm not sure if it's much of a help as most of the errors come from reflections. We have a local differential GPS transmitter on a mountain top, which really improves the accuracy. Unfortunately, it doesn't cover some of the obscure areas I need. I've been thinking of setting up my own DGPS system (yet another project that will never happen). I tried Omnistar correction service:
formatting link
provided amazing accuracy, but costs $800/yr plus hardware. I've also tried a service that differentially corrects locations from recorded data as used by the surveyors (forgot name). That works really nice but also costs money.
Measuring the route trip delay (latency) is NOT the way a GPS works. GPS works by measuring the doppler shift and then using spherical geometry to arrive at a solution.
Well, I'm wondering about his GPS. These days anything more than a few feet _is_ generally done best with a GPS.
Since that's exactly what a GPS does, even if you had the software I can't see it being more precise that the GPS he's already failed to get accurate numbers from.
Let us know if it _does_ happen, the only option I found was very expensive. IIRC, TAPR used to e working on something like that, but I'm not sure it ever happened...
Of course, you're right - there's no round trips to a GPS satellite. However, measuring a doppler shift should be even more accurate. I _still_ can't see how, for any significant distance, he can get more accurate than his GPS.
I have to admit though, that my GPS experience is with marine systems - we never have to worry about the hills :-)
Other than interference from trees, I would expect that your GPS location should be within 20 feet if you let the GPS sit in one spot for a few minutes.
That 800 foot error would have to be due to rugged terrain or user laziness. My confluence in Mongolia was visited a couple of years later, and the second visitor is within a couple of feet of my little pile of rocks, which I put a few feet away from where I meant to put it.
formatting link
relied on the direction arrow, which starts flitting around when you get to the spot, and didn't think to switch to the location display until after I left ;-(
Oh, trees! Your
formatting link
is in an area so heavily covered by trees that a couple of the visitors commented that they couldn't get a GPS lock near the confluence. I'll have to take your word for it that those pictures aren't all of the same tree ;-)
Generally, I rely on the GPS to get me to within visual distance of some desired spot. I was able to locate a gate in a long fence in pitch black darkness with a GPS waypoint. Trying to find the corners of my property worked well, but I was looking for existing markers. Trying to note the exact angle of my house for a solar installation by plotting all four corners leaves me with a rather lumpy looking house.
If you want, I'll post the results of some recordings I've done with VisualGPS software. There are lots of points that are way off the map due to reflections or just plain lousy filtering in the GPS. My Magellan Map 410 really screws up in the presence of reflections. My pile of Novatel Allstar-12 boards that are built into assorted trackers is MUCH better with lots of software control.
The problem is what "within 20ft" really means. Does it mean that the majority (half) of the points lie within a 20 ft radius? Does it mean that the statistical average, or median value are within a 20 ft radius? Does it mean the one sigma distribution of values are within a 20ft radius? Does it mean that ALL the points, except those that are obviously out of range, lie within a 20ft radius?
They aren't. It was an absolute nightmare getting a fix down in the trail below the golf course. Poison oak everywhere. Steep slope. Dense tree canopy. Reflections from the hillside. I dragged in a
30ft fiberglass extension pole to get somewhat into the clear and to pickup the distant differential GPS signal, but still had problems with the tree canopy and the hillside. Taking a series of careful fixes over about 2 hours, using my laptop to crunch the numbers, I located the exact point was on the side of the hill at an almost vertical cliff covered with poison oak. I went back in the winter, when the poison oak is not as prevalent and nearly killed myself trying to place a marker. We didn't have a camera handy on the first expedition, and the digbat in charge of the digital camera had the focus set to "macro" which resulted in all the photos being out of focus. Next time I have a suitable excuse to drag myself up there, I'll take some photos of the setup.
Around here, the survey markers tend to move by themselves. Over the
30+ years I've lived in the neighborhood, we've had numerous property line disputes and several adventures in civil court. It's all part of mountain living. The 1989 Loma Prieta earthquake moved just about everything in the area. Overlaying the 1930's tax maps with the current LADAR surveys show an error of about 35ft. When the county reconciled the PacBell property maps, with the country road maps, and the county tax maps, throwing in the neighboring county's border maps, the results were errors of about 10-50 ft.
Incidentally, the error converting from NAD27 to WGS84 is about 200ft E-W and 4ft N-S in this area. Whenever I an error of exactly this magnitude, I know exactly what happened. I don't wanna even mention the discrepancies between DEM, DLG, DRG, DOQ, SRTM, ad nausium.
Anyway, 20ft radius, by whatever standard, is grossly insufficient for locating a radio tower for propagation studies. I try for about 2 ft radius at most.
A laser range finder is accurate to about +/- 3ft for the consumer variety, 18 inches for the more expensive systems, and errors measured in wavelengths of light for the interferometer variety. That's compared to GPS with about 30ft radius for each end point or a possible +/- 60ft linear error. Up to about 2000ft, I'll take the laser rangefinder.
Also, no reflections, clear sky, no altitude issues, and minimal GPS jamming and interference. For fun, you might wanna try entering the harbor using only the GPS (as if you were sailing in pea soup). Great fun and good for an adrenaline rush.
It died thanks to Motorola getting picky on selling development systems and then obsoleting the Oncore VP boards.
If you wanna do it all in hardware, here's your shopping list of possible boards:
formatting link
I have a pile of the Novatel/CMC Allstar-12 boards. Wanna play with one? That chart says it will do DGPS encode. The catch is that the interface is TTL and splits into 3 serial ports, each of which requires a TTL/RS-232 level shifter. The schematic is simple, but you'll have to build your own interface board.
formatting link
Light reading:
formatting link
Of course, you could do it all post mortem over the internet:
Well, the RS-232 level shifters are easy, so that's awfully tempting, but I've got a lot on my plate at the moment, let me find my Round Tuit first.
Doesn't the 'standard' DGPS use a 50KHz(?) transmitter? I'd like to build something that doesn't need custom data transfer, as the maintenance and user support on VHF data links would probably end up overwhelming me.
That's why my pile of GPS boards is just sitting there. However, when I do decide to do something with them, it will probably be with a proper piggy back circuit board. I'll let you know if I get inspired. We're building APRS trackers in small quantities and it would be nice to get some local accuracy with a local DGPS xmitter.
The only standard is RTCM SC-104 DGPS protocol standard. How it's delivered from the DGPS xmitter to the mobild GPS receiver is not part of the standard. The USCG uses 300KHz beacon xmitters. The airports use some 115Mhz NAVCOM frequency. Omnistar and WAAS deliver via Inmarsat. Surveyors and farm equipment use commercial FM frequencies. Even the data rate is not fixed. Most commonly, the data is localled docoded and delivered via the usual 4800 baud RS-232 ports. If I ever get mine going, it will be similar to the TAPR attempt and use some ham radio simplex VHF/UHF X.25 packet FM channel.
Nope. No time, and in some cases, no programming abilities. Most of the stuff on the page is for DGPS corrections on the receiver end, not transmitting my own DGPS signals. I wish I had known about the DGPS online server at the time, as it would have been useful. My "testing" was part of a consulting project where I was paid to investigate the cost of improving local accuracy. I get less inspired to tinker when I do it for myself, for a hobby, or for entertainment value.
I may try that. To be honest, I just get confused by GPS devices -- they don't seem to speak my language. I bought one at Wal-Mart, got it to work once, and then returned it because I couldn't get anything else functioning the way I wanted it to. Talk about accuracy, I have done this a few times: measured steps.
What I'd like, though, is software that somehow sits at a Ethernet connected PC, then runs on the laptop, and know the exact distance.
Take one GPS. Find one end of the wireless link. Write down the lat-long on a piece of paper. Take a walk to the other end of the wireless link. Write down the lat-long on a piece of paper. Go home and fire up your web browser. Surf to:
formatting link
formatting link
formatting link
formatting link
formatting link
inscribe the lat-long of the end points into the web page. It will calculate the distance for you.
Do you want the straight line distance, as wireless flies? Or the wireline distance as the cable snakes through the landscape?
There was a previous discussion on the relative merits of MSN Virtual Earth location finder:
formatting link
might wanna do some Google searching and see if this is relevent to whatever you're trying to accomplish.
Cabling-Design.com Forums website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.