[R-sig-Geo] optimization with nlm

[Saman Monfared]

Dear All.
I am very new in R.
I am trying to fit a spatio-temporal model on empirical covarance.
I calculate spatio-temporal empirical covarince that is a matrix and
now I want fit a covariance model like :

f <- function(h,u,alpha,phi)
{
exp(-alpha*u*phi)*besselJ(h*phi, nu=1)
}

How can I optimize parameters of a model?
How can I use nlm?
I read an example in help of nlm but I couldnt understand it.
Help me.

-- 

Saman Monfared
Msc Student,
Department of Statistics, Shiraz University,
Shiraz 71454, Iran
Email: Samanmonfared1 at gmail.com

Tel: +98 917 5305167