Accuracy of radio/GPS-like system?

He wants to know which way the handheld unit is pointing WRT some baseline.

they look parallel to me :)

--

Bye.
   Jasen

the only problem I see with this is that the operator holding the device will distort the readings.

--

Bye.
   Jasen

Um, the word "about" which you left out. The phrase "about eight inches square" does not specify that even one edge is straight, let alone that the object is square. An old rag torn from a tee shirt could be described as "about 8 inches square".

As I showed with the diagram, and you recognized below, ambiguity made a 90 degree difference. And it can make a difference between

0 and 180 degrees. The line must be defined, just as you state:

Pick a straight line --

Exactly. I drew the reference line, removing the ambiguity. And it does completely change things - from the ambiguity of no reference line to the precision of a reference line.

Now that is ridiculous. You won't be orienting text in the field. The text represents the handheld unit, as you said. It is not, nor does it represent, a reference line. It is still not completely described, but at least now we know it *is* square. And, since you say we can use an edge as the reference line, we know the edge is straight.

This is what you don't seem to get: *you* are the one who has to specify the orientation. We've (responders to your post) have never seen the hand held unit. We won't be using it. We don't know what it does, other than the function (position, angle) you've requested. Obviously that information is used for something, which for all we know is also incorporated in the unit. For all any of us know, the surface is irregular, making placing an arbitrary line on it difficult or impossible. We're trying to consider the engineering aspects, which, like it or not, requires a clearer definition that you have given. For all we know, there could be some feature on the hand held unit that you want to aim at something.

Um, yes it was. You did not describe a reference line, or state that we could pick any line contained within the perimeter of the object we wanted. Thus the unit remained the equivalent of a mathematical point with respect to determining whether it is parallel to a line, which is the context being discussed.

I have no interest in arguing with you. What has transpired can be netted out to the salient point: you want accuracy beyond what you can get with the approach you have described, in my opinion. To get the accuracy you want, if it is possible within your budget, two things will be needed: a different approach and precise descriptions. You could google on GPS accuracy to gain an understanding of what accuracy to expect from that portion of the setup and function (ie position) and also look into determining the angle optically or via RF phase detection or perhaps some other method to see what kind of accuracy you can get. I also think you are doomed to lower accuracy unless you can lock the hand held unit in place during the measurement period. You have two axis to worry about for determining the angle that the line on the hand held unit has to the line formed by towers A-B or B-C or A-C. You have not specified that the angle in the vertical plane is of no interest, but it *seems* sure that it is not. However, a vertical angle will affect distance, just as a horizontal angle will, and if the angle is to be determined by trigonometry, the distance affects the computation of the angle. We can help you consider some of the factors with which you may not be fully familiar. If you won't consider them when they are brought up, we can't help you.

Ed

The amplitude, yes - but if you're below, say, 175 MHz, the operator's presence shouldn't affect the _phase_ enough to make it unreliable. If you're using microwaves, then all bets are off, and you'd need four directional antennas - and in any case, according to the OP's description, the operator will be just more ground clutter.

Good Luck! Rich

Yes, "8 square inches" is ambiguous, as it doesn't specify the aspect ratio - a ribbon 1/4" wide and 32 inches long has an area of "8 square inches." "8 inches square" removes any ambiguity as to aspect ratio: it's a square, 8 by 8 inches.

Cheers! Rich

hmm, if the antenna is reasonably clear of the ground (or the ground is flat) that direction finding antenna scheme could give you the data needed to determine position and direction

the directions to the three transmitters is enough info to uniquely define a point inside (or outside) the triangle. but I'm not sure how much precision can be had without breaking the bank.

5 degrees would be pretty simple, 1 degree would be aproaching the practical limits afaict.

You may do better with an optical system.

put differently pulsed IR (or visible light) sources on the poles and use a rotating scanner ( to detect and measure the angle to each one...)

Bye. Jasen