Slashdot Mirror
Patent Filed for Underwater GPS
Matthew Sparkes writes "GPS doesn't work underwater, as the signal cannot reach the satellite from a submersible, but researchers have now patented an add-on to the system that could provide GPS navigation for submarines. A base station is tethered to the sea bed at a known depth and GPS location. A submersible anywhere in the area sends out a sonar pulse to which the base station replies with a signal, giving a GPS position and depth as well as the bearing angle from which the submersible's request arrived. The submersible then uses its own depth, which is easily measured, plus the round trip pulse time and the bearing angle sent by the base, to calculate its own position."
This is great!
How long before lost submarines are meandering up our rivers and streams because the GPS mapping told them this was the way to go?
On a slightly more serious note, no self respecting spy submarine will emit a ping to this service ever. There is no way you would want to give your position away so freely.
liqbase
Little nit pick ... GPS signals go from the satellite to the receiver not the other way around.
Men will do anything to avoid stopping and asking for directions
Use your head, can't you, use your head,
You're on earth, there's no cure for that - S. Beckett
Been there, done that. Only difference is we didn't have GPS, only LORAN. You don't need GPS, you just need a sonar transponder whose location is well-known.
1977, aboard RV Melville (Scripps IO). We drop 3 sonobuoy transponders to the ocean floor in a large triangle (few kilometers per side). We know the approximate locations only, since they were after all dropped. Ship sails around doing research and pinging away; record round trip times to each transponder; invert large number of observations to solve for locations of each transponder relative to each other; within a day we know the relative locations accurate to within a few meters (maybe better, I don't recall); meanwhile ship is recording LORAN locations; the LORAN locations are cross-correlated with the relative transponder locations (which are more accurate); net result is that transponder coordinates now have a geographic reference (xy to lat-long).
Two issues with the GPS version: (1) you need to anchor to ocean bottom and have antenna at surface, therefore you need a lot of cable/wire; (2) the surface GPS (antenna) position is NOT the same as the transponder, since the cable is certainly not going to be perfectly vertical. Maybe you don't need to anchor it, just let it drift, then #1 doesn't matter.
Someone said this sounds eminently patentable. No, I don't think so!
I hate it when people are perjuditial to wheels.
Patrick Doyle
I mod down every jackass who puts his moderation policy in his sig. Oh, wait a sec....
This misconception has caused no end of headaches for one company I used to do consulting work for. The company was outfitting fleets of cement trucks with CDPD tracker modems -- these devices had a fully-featured GPS receiver, and could be configured to transmit latitude and longitude to a specific IP address on a specific port at a specified interval. (Nuts-and-bolts: the data was typically sent as UDP packets, so no guarantee of delivery, and CDPD is an older standard for data transmission over digital cell networks, with a max throughput of maybe 64 kbps.) We had software that would aggregate the GPS data for the entire fleet at a server, and then client software which would talk to the server and show real-time reports on the fleet, as well as determine who was at a job site and how long they were there, etc.
Unlike what another poster stated regarding cell phones, the tracker devices we used did all the GPS processing on-board, so what was sent via UDP was either a NMEA string (easily parsed) or some simple proprietary binary format. We would do further corrections at the server to account for various map books and which USGS survey data they were based off of.
Anyway, the problem we has was the truck drivers and their misconception of how GPS worked. Many of the more paranoid truck drivers (and there were a lot of them) were absolutely convinced that we were beaming personal data about the drivers themselves to GPS satellites, forwarding it to who knows where. Trying to explain to these folks that GPS doesn't work that way only resulted in angry confrontations. When I started working on a badging project so that our client could further track the comings-and-goings of the drivers, the hostility and resistance reached alarming levels, to the point where I almost couldn't get work done.
Then again, the whole reason for the software's existence in the first place was to provide documentary proof of the misconduct of drivers. Things like guys taking half-hour naps in their trucks after finishing a job site, or over-slumping their load of concrete so they can sell some excess concrete to a buddy finishing his driveway... We implemented autmated job-site entry and exit discovery because we found that giving drivers a set of pushbuttons to signal when they were starting or stopping a job was just a recipe for abuse. (Funny enough, we kept the pushbuttons to see just how big the discrepancies were between when drivers said they were working and when the GPS claimed they were working. It was eye-opening.)
The drivers were unionized in most cases, so a high standard of proof had to be met. I'm sure that contributed to the air of hostility. But it's also true that many drivers were using fake credentials (many being undocumented immigrants), so the paranoia over a potential loss of privacy and transmission of personal data to a "big bird in the sky" wasn't just because people were worried about getting caught napping on company time.
Not mentioning the names of any companies (nor any specific geographic place names) to avoid legal hassles.