Hello, I'm not quite sure about what you mean. Do you means that you want to use Kalman filter with the output from sensors? If I am right, I think you have to understand the measurement model (how to express output with states estimated by the filter). Then you can use the output as "measurement" and estimate the state by the algorithm of Kalman filter. If your dynamic model is nonlinear, you should use extended Kalman filter instead.

Perhaps I did not understand your question clearly. Contact with me if you have any questions. Thank you. Good luck!

Have you got the model ready? I suspect you need to start with "standard model" for navigation and add TEC as additional state.

Having said so the way you define your model will greatly influence your results as already been mentioned.

If you dont have any kalman filter code so far, I suggest googling for some existing code (there are a few of them around) and starting from there.

Perhaps Matlab codes written by prof Borre will help you (packages Easy and Easy II), http://kom.aau.dk/~borre/.

Let me point you out these sites where one can find several GPS tools in Matlab:

http://www.ngs.noaa.gov/gps-toolbox/index.html

http://kom.aau.dk/~borre/

http://kom.aau.dk/~borre/easy/

http://www.ngs.noaa.gov/gps-toolbox/Borre2.htm

http://kom.aau.dk/~borre/life-l99/