Try this example:

http://rpubs.com/skydome20/93614

Hi Fahad:

I have seen this example, thanks though. I have the source data, I did use the fastICA function, but in the final plot stage [plot(1:1000, a$S[,2], type = "l", xlab = "S'2", ylab = "")], I could not understand the function of of 1:1000. Any suggestion? 

I have six variables and five sampling locations.

Hi Subhronil,

1:1000 will generate 1000 points or samples in x axis for each sample in the data set, which are five in your case. Try to change the function to 1:5 and see if this works?

Hi Fahad:

Should I change the function to 1:5 or to the total number of data points? I have 145 data points, and I did change the function to 1:145, and got this:

Centering

Whitening

Symmetric FastICA using logcosh approx. to neg-entropy function

Iteration 1 tol = 0.00491022

Iteration 2 tol = 0.0001445918

Iteration 3 tol = 6.99492e-06

I also got 2 bivariate graphs, each one for S'1 and S'2.

Any suggestion?