Cross Correlation to determine Delay

Hi,

I have 2 signals of equal length. The signals correlate but with a time delay. I thought of doing a cross-correlation between both signals to determine the lag at which both signals have the maximum correlation. I've written the following line of code to do that.

[c,lags]=xcorr(signal1,signal2,20);

Plotting both signals and their cross-correlation returns this.

I was expecting to see a peak at approx. positive or negative 10 because that's when both signals drop significantly. I can't seem to figure out why that isn't the case. Is it because both signals have the same length?

Any help would be very much appreciated!

[Tim]

Both of the signals have a large bias; your cross correlation values are going to predominantly be affected by this bias, not signal variation. Try making the signals zero-mean before cross-correlating.

[Samuel]

Can I achieve that by subtracting the mean from my signals?

[Tim]

yes