Can an Extended Kalman Filter be used on a noisy Sinewave

[Ultimâ]

If yes, what model is used? For example a moving target could have a state of
\[\underline{x}_{k+1}=[x_k, y_k, \dot{x}_k, \dot{y}_k]^T\]
...? I just can't get my head around this. Any assisstance would be great.

 

[LouisSheffield]

Re: Can an Extended Kalman Filter be used on a noisy Sinewav

I'm going to read this one too, so I can't say if it's precisely what you're looking for.
However, it (or its references) should be very close.


**broken link removed**


There are many covariance based (Capon and APES, or eigenvalue methods) methods that perform quite well. In your exploration, you'll find that maximum likelihood performs to a much lower SNR than many other methods (Kay, Tretter, etc).

Then there's the FFT, which will have the best noise mitigation for a stationary sinusoid. If nothing else, FFT results serve as a great "starting point" for developing a model to constrain the behavior of the various stochastic filters.