| Description |
This script generates artificial spatial data using a first order spatial autoregressive process (AR1)
The process is
X(i,j) = PHI*(X(i,j-1)+X(i,j+1)+X(i-1,j)+X(i+1,j) + error
To generate a 10x10, periodic spatial pattern, with normal error distribution, and a variance of SIGMA, the command is
spatialPatternAR1([10,10],PHI,SIGMA)
or equivalently,
spatialPatternAR1([10,10],PHI,SIGMA,1)
the same pattern, but with a rigid boundary is generated with
spatialPatternAR1([10,10],PHI,SIGMA,0)
Different error distributions can be used. Using the current script this can be done by extracting the generator matrix, M
M = spatialPatternAR1([10,10],PHI,SIGMA,1,1)
M is a 100x100 matrix.
Then given a random vector, error (size = 100x1) with indepedent elements sampled from the required error distribution, the spatial pattern is given by
x = reshape(M * error,[10,10]) |
