System Identification via Polynomial Transformation Method
This work addresses the problem of system identification for linear systems, offering a method that handles general sampling and noise conditions, though the improvement is incremental.
The paper proposes a polynomial transformation method for extracting system poles from impulse response data under general sampling and noise conditions, demonstrating superiority over two existing methods in uniform sampling cases.
We propose a method based on minimum-variance polynomial approximation to extract system poles from a data set of samples of the impulse response of a linear system. The method is capable of handling the problem under general conditions of sampling and noise characteristics. The superiority of the proposed method is demonstrated by statistical comparison of its performance with the performances of two exiting methods in the special case of uniform sampling.