error: matrix dimensions must agree

Hi

I am banging my head over this least square curve fitting even when following the simplest procedure I found. The error says,

"Error using /

Matrix dimensions must agree." - in line where the function 'func' is defined. Anyone have answers?

Thanks

clc 
close all
clear
background = csvread(['C:\Users\shadh\Downloads\vnaSweep\ar2000mT\5.5mm length\0.1mmReCal\0w\0SM.csv'],3,0,[3,0,1603,2]);
backgroundImpIm = background(:,3);
f = background(:,1);
freq = f(100:1601);
x0 = [1,1];
h = 5E-3;
a = 5E-5;
m_e = 9.1E-31;
q = 1.6E-19;
c = 3E8;
omega = 2*pi*freq;
k_0 = omega/c;
beta = k_0;
epsilon_real = 1;
epsilon_im = 0;
x = epsilon_im./epsilon_real;
func = @(K_a,freq)(-K_a./(tan(2*pi*freq*h*(1 + 0.19/((K_a/60 +1) - 0.81))/c)));
K_a = lsqcurvefit(func,x0,freq,backgroundImpIm(100:1601))

[KSSV]

ONly one csv file is uploaded....error line is not mentioned.

[sxh]

Sorry about that. I just edited the code..

Answer 1

[Torsten]

K_a in your function definition is a scalar. Thus you have to change x0 to be a scalar, too.

clc 
close all
clear
background = csvread(['0SM.CSV'],3,0,[3,0,1603,2]);
backgroundImpIm = background(:,3);
f = background(:,1);
freq = f(100:1601);
x0 = 1;
h = 5E-3;
a = 5E-5;
m_e = 9.1E-31;
q = 1.6E-19;
c = 3E8;
omega = 2*pi*freq;
k_0 = omega/c;
beta = k_0;
epsilon_real = 1;
epsilon_im = 0;
x = epsilon_im./epsilon_real;
func = @(K_a,freq)(-K_a./(tan(2*pi*freq*h.*(1 + 0.19./((K_a/60 +1) - 0.81))/c)));
K_a = lsqcurvefit(func,x0,freq,backgroundImpIm(100:1601))

Local minimum possible.

lsqcurvefit stopped because the final change in the sum of squares relative to 
its initial value is less than the value of the function tolerance.

K_a = 48.3478

[sxh]

Youve got good eyes mate. Thanks. Spent almost a full day for the error, wonder how I missed this one.

Thanks a bunch,

S

Answer 2

[James Tursa]

I would presume you may need element-wise operators. Try this:

func = @(K_a,freq)(-K_a./(tan(2*pi*freq*h.*(1 + 0.19./((K_a/60 +1) - 0.81))/c)));

[sxh]

I tried that too. Now the error is in the next line:

"Error using lsqcurvefit

Function value and YDATA sizes are not equal.

Error in ka2 (line 26)

K_a = lsqcurvefit(func,x0,freq,backgroundImpIm(100:1601))"

[James Tursa]

Type the following at the command line:

dbstop if error

Then run your code. When the error happens, the program will pause with all variables intact. Examine them to figure out which variables are causing the dimension problem, then backtrack in your code to figure out why the dimensions are not what you expected.

[sxh]

The only culprit I can think of is 'K_a' but this is the one I am trying to solve for.

I replaced 'K_a' with just a scalar value 1 to see if the function value the same as the size of the YDATA. They are the same size.