信号相关性
您好，我有两个复杂的函数（大小 1x1x2501），我需要在它们之间进行关联（互相关）。
我尝试使用此命令：
[c,lags] = xcorr(Hmimo_tb(1,:)',Hmimo_tb1(1,:)','coeff');
其中 Hmimo_tb 和 Hmimo_tb1 是我的两个信号，其中唯一的区别是它们是在不同位置测量的。这两个信号之间的差异最大等于1.5e-13，因此它们仅受噪声影响。
当我使用 xcorr 并去分析“c”时，我发现它很复杂，所以我这样做：
abs(c)
通过这种方式，我拥有所有实际价值，在滞后 0 周围标准化。 我的问题是我必须构造一个具有相关系数的矩阵。我的意思是我应该有这样的东西：
M=[1 corr(Hmimo_tb,Hmimo_tb1); corr(Hmimo_tb1,Hmimo_tb) 1]
其中 1 是矩阵上侧第一个的 corr (Hmimo_tb,Hmimo_tb) ，另一个是 cor(Hmimo_tb1,Hmimo_tb1) 。
使用 xcorr 命令我在分析结果时遇到了巨大的问题。首先我发现只有一个值 1 和很多非常接近 0 的值...这对我来说很奇怪，因为我预计所有值都在 0.8 左右、0.9 或 1...相反，我发现了很多像 0.001 或接近这个值的值，为什么？我该如何解释这些值？ 从理论上我知道相关参数定义为 -1<=p<=1 但是绘图中差异较小的两个函数（我说的是 1.5e-13）怎么可能不相关呢？
希望能找到快速的帮助，谢谢大家。
Hello i've two complex functions ( size 1x1x2501) and i need to do a correlation between these (cross-correlation).
i've tried to use this command:
[c,lags] = xcorr(Hmimo_tb(1,:)',Hmimo_tb1(1,:)','coeff');
where Hmimo_tb and Hmimo_tb1 are my two signals in which the only difference is the fact that they have been measured in different positions. The difference betweeen these two signals is max equal to 1.5e-13, so they are only affected by noise.
when i use xcorr and i go to analyze the 'c' i find that it is complex so i do:
abs(c)
in this way i have all real value, normalized around the lags 0.
My problem is that i have to construct a matrix that has the correlation coefficients. I mean i should have something like this:
M=[1 corr(Hmimo_tb,Hmimo_tb1); corr(Hmimo_tb1,Hmimo_tb) 1]
where 1 is the corr (Hmimo_tb,Hmimo_tb) for the first on the upper side of the matrix and cor(Hmimo_tb1,Hmimo_tb1) for the other one.
Using the xcorr command i have huge problems in the analysis of the results.First of all i find only one value 1 and a lot of value very close to 0...and this is strange for me since i expected all values around the 0.8, 0.9 or 1....instead i found a lot of values like 0.001 or near this value why? how can i interpretate these values?
from theory i know that the correlation parameters are defined as -1<=p<=1 but how is it possible that two functions with a low difference in the plot (i'm talking about 1.5e-13) are not correlated?
Hoping to find a fast help i thank you all.
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（1）
事实上，您有一个“1”值，而互相关矩阵的其余部分接近于零，这表明互相关性非常强。这些值通过两个输入的幂进行归一化，因此不会在输出中得到“1”。事实上，这表明互相关接近完美。 “接近 0”表明它们在其他阶段没有很强的相关性。
The fact that you have one "1" value and the rest of the cross-correlation matrix is near zero indicates a very strong cross-correlation. The values are normalized by the power of the two inputs, so it is not given that you will get a "1" in the output. In fact, that indicates that the cross-correlation is nearly perfect. The "near 0's" indicate that they are not strongly correlated at the other phases.