Kalman filter design for impedance measurement data

[north2012]

Hello,

I am trying to implement Kalman fiter to remove some Gaussian white noise. I have succesfully implemented moving average filter but I would like to compare them in Matlab.

Lets assume that I have nominal values for model parameters:

Re = 917.5;
Ri = 629.0;
Cm = 3.42e-9;

while model is described with

Z = Re*(1+1i*w*Ri*Cm)./(1+1i*w*(Re+Ri)*Cm);

where

w = 2*pi*f;

and

f = linspace(10e3, 100e3, 100);

Then I would like to add some noise:

rng(0,'twister');
a = -1;
b = 1;
r = (b-a).*rand(100,1) + a;

Can anyone help me with Kalman filter code for white noise cancellation?

Thanks in advance.

 

[KlausST]

Hi,

I can't help you with kalman filter.

I'd use a single frequency DFT.

Klaus

 

[ahme0307]

Hi,
To Model your model with Kalman filter, You will need state transition matrix and variance of measurement and process noises.
If you are looking only for comparison purpose, Kalman filter is already implemented in matlab package.
Check this link:
https://se.mathworks.com/help/control/examples/kalman-filter-design.html

 

[north2012]

Thank you very much, but how to recognize matrices A, B, C and D from my data?

Thanks...