lattice representation

[Paul Box]

Hi John

For a first look, you might try "Applications of Spatial Data
Structures" by Hanan Samet (Addison-Wesley).  It has a whole chapter
dedicated to neighbor finding techniques as tree traversals.  That may
not be the index you're looking for, but it might be more efficient
than what you're using.  I'll be interested to hear how well this
addresses your problem-- I'm still using the directional lookup that
you describe.

John Carnahan writes:
 > 
 > Does anyone know any good references on lattice/grid representation? My
 > question is how could I best represent a 2-dimensional lattice with
 > neighborhood information. I am most interested in storage and retrieval of
 > neighbors from a linear table (eg SQL). Storing coordinates and then doing
 > directional lookups (N,S,E,W ...) cannot be the best way. 
 > 
 > For example, let's say that I have a given list of values/cells indexed by
 > an ordered set of integers. Then for a 1-dimensional lattice all neighbors
 > of cell x are given by indices x+1 and x-1. For a 2-dimensional lattice
 > things aren't so simple because any cell x has several neighbors (eg
 > hex=6, rect=8). There should be some way to index a set of values for a
 > 2-dimensional lattice such that the position (or index) of the neighbors
 > of a given cell are described by its index. 
 > 
 > There must be tons of material written on this especially in the CA and
 > GIS literature. I thought this might be the best forum in which to ask
 > although it is not specifically related SWARM. 
 > 
 > Thanks,
 > 
 > JohnC
 > UCLA
 > 
 > 
 > 
 >                   ==================================
 >    Swarm-Modelling is for discussion of Simulation and Modelling techniques
 >    esp. using Swarm.  For list administration needs (esp. [un]subscribing),
 >    please send a message to <address@hidden> with "help" in the
 >    body of the message.
 >                   ==================================
 > 
-- 
//////////////////////////////////////////////////
// Dr. Paul Box
// Dept of Geography and Earth Resources
// UMC 5240
// Utah State University      http://www.nr.usu.edu/~sanduku
// Logan, UT  84321           Phone: 435-797-2546
//////////////////////////////////////////////////

*  If Elvis was so great, why was he buried in the back yard like a
*  hamster?


                  ==================================
   Swarm-Modelling is for discussion of Simulation and Modelling techniques
   esp. using Swarm.  For list administration needs (esp. [un]subscribing),
   please send a message to <address@hidden> with "help" in the
   body of the message.
                  ==================================