Implementation of Discrete Kalman Filter in MATLAB

[beginner_EDA]

Hello Everybody,

I have set of 100 data points = [ 400, 405, 408,.........,420] and I want to use kalman filter on it.

Unfortunately, I don't have any more information for state equation, output equation, process noise, measurement noise, matrix A, B, C, etc.

How could I proceed ?

 

[rakue]

I actually have the same question, I've read that it is possible to develop the algorithm based on a set of measurements

https://www.ifcs-eftf2011.org/sites/ifcs-eftf2011.org/files/editor-files/Slides_Grewal.pdf (see slide #7)

Does this mean that it's possible to develop a filter based on the raw data or do I still need a function that will describe the data? Because if it's the latter, then what I would suggest to the original poster is to first find a function that will describe your data, let's say you used least squares regression in order to come up with a function. And then from that function, you can apply it using the filter.

I'm still in the process of understanding the Kalman filter myself, so if there's another way to develop a filter without knowing the state equation, then I'd like to know how as well.