Fitting a vector to another

5 次查看(过去 30 天)
Asl
Asl 2014-7-8
I have two vectors x1 and x2 (2 data sets) of two sensors measuring the same value (redundancy measurement). Correlation is almost 0.99). I want to adjust one vector (x1) to another (x2) (using other as a reference). I am currently doing 1x/x2 (simple normalization). Any other options? Thanks
  2 个评论
Image Analyst
Image Analyst 2014-7-8
Give a small example. Let's say x1=[1,2,3] and x2 = [1.110, 1.999, and 3.042]. What do you want as output?
Asl
Asl 2014-7-8
I have to use the reading of X1 for some reasons but X2 is more relaibale and accurate. Thus I am normalizinh x1 by dividing it is x2 to fit X1 according to the X2.

请先登录,再进行评论。

回答(1 个)

John D'Errico
John D'Errico 2014-7-8
编辑:John D'Errico 2014-7-8
You are the one, the only one, who can make a decision like this.
You have posed an implicit model:
x1 = k*x2 + e
where k is an unknown parameter to be found in your calibration scheme, and e is a vector of errors.
So the question is what model do YOU think makes sense? Then we can see how to estimate it. For example, might there be a constant offset on one axis or the other?

类别

Help CenterFile Exchange 中查找有关 Interpolation 的更多信息

标签

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by