Calculating the distance between 2 points 1

[WhistlingMountain]

I'm much more of a software person, and in general this project is very much software heavy, but built on a hardware idea I'm not very educated on.

I'm looking for the types of available technology out there that can fit this problem.

A system in which a sensor can measure and store the distance of a bunch of receivers as the receivers move around in an area no further than 150 yards from the sensor.

The receivers must be small and basically weightless, but not nano size, just micro size. quarter to dime.

It must be able to work without LOS, but it in an open area, so no walls or anything.

It only has to be "relatively" accurate. Probably within a yard for a pre-alpha version.

For now I don't really care about cost, mostly availability as I'm just scoping things and researching.

I am just looking for technologies and articles about them, basically any pointing in the right direction anyone can offer so I can get better educated. Thanks for any help!

 

[IRstuff]

Does it have to be in real-time?

Kinematic GPS-capable receivers can do this real-time.

Higher-end GPS receivers can also do this, but not real-time.

Differential GPS



TTFN

FAQ731-376

 

[WhistlingMountain]

It does not need to be displayed in real-time

If the movements happen and then 10 minutes late I have the data that is fine. I just meant it wasn't a stop and wait type thing, the receiver will be constantly moving so the data must be constantly collected while it's happening.

Isn't GPS overkill? Also I thought even with the higher end GPS it was still only within 7 feet or so.

There's nothing local that can do it better/more efficiently?

I'll checkout the Kinematic GPS, Thanks

 

[WhistlingMountain]

I've stumbled across WiFi tags, they look to be a pretty good step towards what I'm looking forward, unfortunately looks like they're still kind of big. Anyone had any experience with them?

 

[VEBill]

At least for the aspect of the project that you've described, it's the software that is very nearly trivial and the hardware architecture and device selection that will probably be THE primary design issue.

Can you turn the system around so that the mobile units are the transmitters and the receivers are fixed? Your description of the non-realtime nature of the requirements implies that you could. It would probably be easier.

 

[IRstuff]

Seems to me that the "project" is probably not about developing object locators. But, if you are rolling your own, then it will become a "project" on its own, with its own issues and schedules unrelated to your primary "project."

Your time scale is awkward, since the postprocessing GPS' I'm thinking of are more like a day.

Real-time differential GPS can achieve positioning better than 1 cm, depending on the brand. There are some differential-capable GPS' that run about $800 each, but require something like a radio-modem to feed the differential corrections from your reference station.

The main advantage of GPS is that much of the infrastructure is already done and debugged for you.

TTFN

FAQ731-376

 

[WhistlingMountain]

"At least for the aspect of the project that you've described, it's the software that is very nearly trivial and the hardware architecture and device selection that will probably be THE primary design issue."

The project is 90% software in processing the desired simple distance data, however, I am a software person and already know how to write most of the algorithms needed.

"Can you turn the system around so that the mobile units are the transmitters and the receivers are fixed? Your description of the non-realtime nature of the requirements implies that you could. It would probably be easier."

Yes, they can be the transmitter.

"Seems to me that the "project" is probably not about developing object locators. But, if you are rolling your own, then it will become a "project" on its own, with its own issues and schedules unrelated to your primary "project.""

Correct I do not want to develop my own transmitter/receivers.

 

[MacGyverS2000]

Sounds like a school project...

Dan - Owner

http://www.Hi-TecDesigns.com

 

[WhistlingMountain]

I'm 5 years out of college, and work as software developer...

So, no, not a school project.

 

[VEBill]

There have been Eng-Tips threads on similar projects before. I don't recall any of them getting a satisfactory solution.

 

[IRstuff]

That's because EVERYONE thinks that this is a simple problem to solve with own-built hardware, but it's not.

And NOBODY wants to spend the money to buy the DGPS or RTK hardware, even though it's still cheaper than building your own system.

TTFN

FAQ731-376

 

[WhistlingMountain]

yea I'm starting to notice this. I think the best solution I've found thus far have been WiFi tags.

You know you're not going to find much when searches for what you want online return a bunch of patents and not a bunch of products.

 

[WhistlingMountain]

"And NOBODY wants to spend the money to buy the DGPS or RTK hardware, even though it's still cheaper than building your own system. "

Using satellite really don't fit into this application though.

I'll probably just see what I can do with wi-fi tags for an early version, too big for all aspects of the application, but what can ya do.

I'm surprised there's no simple transmitter-receiver via radio signal hardware pre-existing.

Thanks for the info everyone.

 

[IRstuff]

"no simple transmitter-receiver"

That's because it's NOT a simple problem. 1-m positioning accuracy requires about 1-ns time resolution. To propagate that time and have adequate synchronization to the master clock is PRECISELY what the GPS PRN codes are designed for.

RF direction finding for your specific problem requires angular resolution of better than 5 mrad, that requires a 20-inch antenna at 200 GHz, which is 80 times higher than the 2.4 GHz WiFi frequency.

While these guys:

http://www.ekahau.com/?id=4999

claim 1-m positioning accuracy, it's not at all clear how they achieve that. Moreover, note the price of the developer kit is $5990, so it's not exactly cheap.

TTFN

FAQ731-376

 

[VEBill]

If the mobile units would ping, then you could have several receivers triangulate on them based on time-of-flight. The receivers could also do them same thing amongst themselves to stay synchronized. Maybe use Barker Codes to narrow the pulse and improve SNR. But anyway you look at it, it's a very big development effort. Especially if you want the mobile units to be smaller than the battery inside them (requires bending the space-time fabric).

 

[WhistlingMountain]

What there's no simple way to bend the space-time fabric?

 

[BOMBOVA]

chirp ultrasonic goes for 30 meters or so depending on frequency / power / sensor size. sync up a few from a sensor and track them. GerwingR