issue on cross correlation

Issam Bou

2015 6 8

0 个回答

3 次查看（30 天）

0 个投票

Hello; I'm working on temperature data taken from a pt100 on matlab

I draw the temperature versus time ( I HAVE 11 DIFFERENT SENSORS PLACED ON DIFFERENT SPOTS SO I HAVE 11 DIFFERENT PLOTS ) AS SHOWN BELLOW

NOW I WANTED TO ANALYSE MY DATA USINE CROSS AND AUTO CORRELATION, AND THE ISSUE I HAVE WHEN I CROSS CORRELATE SENSOR NUMBER 2 AND FOR FOR EXAMPLE USING THIS CODE

function interco(x,y)
[X,Y] = xcorr(x,y)
plot(Y,X)

IS THAT I GET A TRAIANGLE MEANING THAT MY DATA ARE IDENTICAL WHICH CLEARLY IS NOT THE CASE

CAN ANYONE HELP PLEASE FIND AN ANSWER TO THIS PROBLEM

THANK YOU

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

Adam 2015-6-8

编辑：Adam 2015-6-8

That doesn't imply they are identical. The correlation is not normalised so a correlation of that magnitude ( 5x10^7) could mean anything in terms of how good the correlation truly is.

All that tells you is that the correlation at 0 lag is highest (as you would expect) and that it drops off rapidly as you move one signal to the left or right of the other.

请先登录，再进行评论。

请先登录，再回答此问题。

Follow Question