Next: Autocorrelation of a noised Up: Correlation Previous: Introduction to correlation Contents

Correlation of noise

We are going to obtain the autocorrelation of `noise'.


-->xbasc()
 
 
-->corrnoise = corr (noise,1000);
 
-->xsetech([0,0,1,1/2])
 
-->plot2d(t(1:500),noise(1:500));
 
-->xsetech([0,1/2,1,1/2])
 
-->plot2d(t(1:500),corrnoise(1:500))

**Figure 9.3:** *Autocorrelation of `noise'. The upper trace plots the signal, the lower one its autocorrelation.*
$\begin{figure} \epsfbox{figures/corr3.eps} \end{figure}$

This is a very important result. The autocorrelation of white noise is 0 except at the origin whose value represents the variance of the signal. In our noise the variance is more or less `4' since our signal has a standard deviation of `2'.

At this moment we can study the effect of noise on the correlation of our primitive signal.

je 2006-10-13