Slashdot Mirror
Military Tech: GPS and Networking
king of birds writes "The New York Times has an interesting article on the present military use of GPS. While some units have rather modern system that can graphically display locations of other troops, others rely on 10-year-old 5 channel receivers. Kind of odd when I can 12 channels on my civilian model (with admittedly lower spatial accuracy)." aaronvegh writes "From the Canadian Press, a story about how a US infantry division uses a system of transponders and servers to track friendly and enemy units, from the headquarters to inside individual tanks. Talk about total information awareness! No friendlies were harmed in the making of this story."
Isn't it a little risky to put location transponders on all your military units? If the enemy cracks your transponder codes, they can easily target you.
Kind of odd when I can 12 channels on my civilian model
Of course, your civilian model probably fails 1% of the time, and wouldn't survive a day in a sandstorm, in part due to it's fragile electronics.
The Military version, while only 5 channels, is probably much more robust then your puny little civilian model.
Well that's why you set up a way point of your home base/camp site/vehicle. When you get lost, you tell it to return to here, and you get a pretty little arrow that says "home is 3 miles that way" and the arrow constantly is updated on the fly. It doesn't just list friggin longitude and latitude numbers..well, maybe yours does.
Finally, math books without any of that base 6 crap in them.
Or, even easyer:
Just crack that, and don't waste any of your precious ammo...
Day-to-day operations give you the chance to test the system out under a real load, so you know it'll work in an emergency.
The biggest problem I've seen is that the channel is just overloaded in many places. People using excessively long paths, too many high digipeaters too close together, and too-frequent transmissions.
OpenTrac doesn't fix that (there are methods already being discussed and tested for those problems), but it does let you do all sorts of nifty stuff with telemetry and such. And it's far easier to encode and decode than APRS, so it's better suited for small microcontrollers.
When all you have is a hammer, everything looks like a skull.
As the farmer said when asked for directions: "If I was you and trying to get there, I wouldn't start from here."
One line blog. I hear that they're called Twitters now.
There are many methods of navigation...
Most are RF Based
VOR (VHF omindriectional radio-range)
ODB (omni direction becon)
DME (distance mesuring equipment)
Loran
Radar can be used for navigation
GPS everyone is familar with.
there are also visual methods
The stars
Dead Reconing, bast on land marks.
Our marines are trained in navigation by stars and I believe they still check their ship's position every knight and compare it to the GPS info.
All pilots are traind Dead Reconing first, as this is a skill most people have (although some haven't devoloped it very well). Soon after they move into Triangulation of radio signals, and eventually you are allowed to use a GPS
Also, if the frequency is changing at a rate of 10kHz, simply doing a Fourier Transform of the signal probably won't help you much in trying to determine the true frequency at any given point in time, especially if you don't have a clue what the frequency changing algorithm or key values are.
My guess is that it's an X11 Unix application being displayed on a windows box running Exceed to make it into a virtual X terminal.
(Yes Virginia, the dumb terminal is alive and well.)
Said configuration is so common it's almost obscene. My first Job out of college was at one of Lockheed Martin's many branches. All of the REAL work was done on various flavours of Unix (AIX, HP-UX and some other IBM OS in our case, and some projects in the facility were expreimenting with Linux and BSD as alternatives (Main problem being, VA and the like don't exactly build their boxen to MILSPEC, HP and IBM were happy to do so.) Obviously, we needed a Unix environment to program computers that would be rinning Unix in the field. Makes sence, right?
Problem being, as they said on Star Trek: "The buerocratic mentality is the only constant in the universe". And LMCO has a BIG one. Some big muckety-muck, a CIO, an IT director, or somesuch, had chosen Dell as the desktop vendor for our facility, gotten several score truckloads of the things at bulk rate, built an NT-centric IT staff and 'standard desktop configuration', and said "Thou shalt use windows on thy desktop!". No matter that windows is completely useless to engineers. He's got his Dell/windows empire, and he's going to lord over it. So what we had to do, is run Exceed on the things to open virtual X windows onto the real computers, on which our actual work was done. This was supposedly a pretty common situation at the rest of LMCO as well.
In the course of doing latter jobs, and interviewing for others, I've discovered that this is stupidly common within other government contractors as well, and not uncommon outside. So I've little doubt that it's pretty common in the actual military as well.
I can't even BEGIN to imagine just HOW many windows PCs are out there, complete with Word, Excel, Powerpoint, Outlook, etc. etc. etc., all those licenses doing nothing but burning money; when the only purpose they wind up serving is as a glorified dumb terminal.
(PS. Oh yeah... it's not too hard to change the graphic on the start menu button, startup screen, or most other places, so that's no indication that it's not windows.)
cya,
john
Imagine all the people...
A few comments:
1) Four satellites are needed for a 3D position lock because the "fourth unknown" is exact GPS time. Receivers simply set up and solve a system of equations with four variables: x, y, z and "b", the receiver's clock bias. Solving the system of equations is an iterative and approximate process.
2) Modern receivers will use as many satellites as are "visible" to them at any point in time. More sats means higher accuracy. Each satellite gets weighted based on its elevation - satellites close to the horizon hardly figure into the calculation.
3) It is perhaps true that a 2D position can be achieved with just two satellites, if the third "intermittent" signal can provide clock (GPS time) calibration at sufficient intervals. Realize though that the accuracy will be SEVERELY limited since the receiver's clocks are vastly inaccurate and temperature-sensitive at the required time-scales. Furthermore, the position is actually a "circle" (intersection of two spheres) which, when additionally intersected with the earth's surface, will give you two possible positions. Usually one can be discounted based on other information. Each of these steps greatly reduces the accuracy.
3) Biggest problem for the military is jamming (NOT SPOOFING) of the GPS signals. The GPS signal level is WAY BELOW NOISE, i.e. a very weak radio signal, and can easily be jammed across wide areas with low-powered jammers. Reportedly one such jammer operated in central Baghdad during the past weeks, causing some "smart" bombs to become "dumb". Reports that the jammer was located within the Russian embassy compound are as yet unfounded. The military uses additional GPS signals on a different frequency band, which makes jamming a bit more difficult.
4) Second biggest problem is also related to the low signal-strength of GPS: receivers generally can get a good position fix anywhere in the open. Inside structures, or under dense foliage, however, GPS devices can become useless. Next generation GPS satellites will be broadcasting on two additional frequency-bands and at a higher signal-strength.
From what I understand, the older military receivers may be very robust, but in past years a number of advancements have been made in GPS receiver technology, to improve on the points listed above and generally allow more accurate position fixes under more adverse conditions.
Another important note: for many applications (including the one detailed in this story), relative positioning would due, or even be more appropriate. Using differential double-differencing GPS, the position of each friendly fighting vehicle can be determined relative to, say, the main comms vehicle. If in addition the position of the comms vehicle is well-known (using a highest-quality GPS receiver for absolute positioning; and preferably by keeping its location fixed for 15 minutes or longer, to allow for time-averaging out certain errors), then you will know the fighting vehicle's positions with MUCH GREATER accuracy then what's possible using their receivers for absolute positioning only.
Such "double-differencing" relative positioning requires a data-link, and since this application already relies on this, it seems like a no-brainer. Maybe they already do this?
Thoughts?
Are you sure that any of the APRS problems are related to the message format?
If not, it is very unwise to create a new, incompatible, message format "just because this one is nicer".
The average ham radio operator considers something like a PIC or a TNC an "investment for life" and will not be prepared to build or construct something new to do essentially the same as what he already has, even if it does it in a better or cleaner way!
This probably is because many hams are just trying out different things, and that of course is the main purpose of the hobby. So, many people will go on APRS just to see what it brings them, not to follow the state of the art or to keep improving things "because it can be done better".
I think that you will find that even when Opentrac is 20 times better than APRS (which I have not yet seen from any of the documents on the site), in 10 years time still most people use APRS. It was there first, it has been implemented on many platforms, and it is in use by a large group of people that have no urge to change. It is like IP version 4. Sure, IP version 6 is nicer but nobody is going to use it as version 4 does the job fine.
Of course you are right about little open hardware and software being available for APRS. But as long as there are no legal problems with that (are there?) it would seem to be better to just work on solving THAT problem (i.e. design and publish an open source APRS device), rather than re-invent the wheel and find yourself largely ignored by the userbase.
Rob