V&H Coordinates?
Alta User
[ Author was Alta Computer ]
[ Posted on 22 Aug 1994 23:09:54 -0700 ]
Can anyone out there tell me how to use the V&H coordinates for CO's and
Tandems commonly used in LERG files, etc. to map Tandems to a to-scale
map of the United States? I have the formulae for calculating distances
between two sets of coords, but until recently I never needed to know how
to map one set of coordinates to the globe.. Any help or leads would be
appreciated. Thanks.
Doug Frank
Alta Systems
Rob Lesieur
: [ Article crossposted from alt.dcom.telecom ]
The coordinates are used as two of the three data points forming a right
triangle, allowing you to apply the pythagorian (sp) theorem.
Bottom line: square root ( (|X1 - X2|) + (|Y1 - Y2|) ) = distance between the
two coordinates. You could use them to map a single point knowing the physical
location of coordinate (0,0).
--
Rob Lesieur
ro...@cbis.com
=> My opinions and statements do not necessarily <=
=> reflect the opinions or policies of my employer <=
Stu Jeffery
>Tandems commonly used in LERG files, etc. to map Tandems to a to-scale
>map of the United States? I have the formulae for calculating distances
>between two sets of coords, but until recently I never needed to know how
>to map one set of coordinates to the globe.. Any help or leads would be
>appreciated. Thanks.
> Doug Frank
> Alta Systems
Some time ago I picked up from Tom Libert the source code for a program
that translates Lat and Longtidue into V and H, but it doesn't go the other
way. You can see from the comments that going the other way is hard. I will
post the code, if there is interest (it is about 8k); or you could get it
directly from Tom Libert
If someone has a program that goes from V and H to Lat and Long, I sure
would like a copy.
/*
* ll_to_vh.c -- computes Bellcore/AT&T V & H (vertical and horizontal)
* coordinates from latitude and longitude. Used primarily by
* local exchange carriers (LEC's) to compute the V & H coordinates
* for wire centers.
*
* To compile: cc -o ll_to_vh ll_to_vh.c -lm
*
* This is an implementation of the Donald Elliptical Projection,
* a Two-Point Equidistant projection developed by Jay K. Donald
* of AT&T in 1956 to establish long-distance telephone rates.
* (ref: "V-H Coordinate Rediscovered", Eric K. Grimmelmann, Bell
* Labs Tech. Memo, 9/80. (References Jay Donald notes of Jan 17, 1957.))
* Ashok Ingle of Bellcore also wrote an internal memo on the subject.
*
* The projection is specially modified for the ellipsoid and
* is confined to the United States and southern Canada.
*
* Derived from a program obtained from an anonymous author
* within Bellcore by way of the National Exchange Carrier
* Association. Cleaned up and improved a bit by
* Tom Libert (t...@comsol.com, lib...@citi.umich.edu).
*
* CASH REWARD for copies of the reference papers, or for an
* efficient (non-iterative) inverse for this program! (i.e.
* a program to compute lat & long from V & H).
*/
sample run:
Computes V&H (vertical and horizontal) coordinates
given latitude and longitude.
Example (Ann Arbor, MI):
Enter location: 421700N0834445W
v = 5602, h = 2919
Enter location: 372326N1221605W (Mt View, CA.)
v = 8577, h = 8686
Enter location: 383517N1212923W (Sacramento, CA.)
v = 8301, h = 8582
Enter location:
-------------------------------------------------------------------------
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
| Stu Jeffery Internet: s...@shell.portal.com |
| 1072 Seena Ave. voice: 415-966-8199 |
| Los Altos, CA. 94024 fax: 415-966-8456 |
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Jim....@umich.edu
Some time ago I picked up from Tom Libert the source code for a program
that translates Lat and Longtidue into V and H, but it doesn't go the other
way. You can see from the comments that going the other way is hard.
It's hard to invert the function algorithmicly, but in practice it's no
problem to do the inverse transformation.
You just have to make an initial guess for lat/long, transform to V&H,
refine the guess, and iterate. Three or four iterations will do it.
Tom Libert is no longer here at CITI.