A new system identification algorithm based on iterative algebraic filtering

A new algorithm for linear system identification is proposed. Unlike most existing system identification algorithms, this approach uses iterative algebraic filtering to enhance the signal in a high-noise environment. The iterative algebraic filtering approach, which is adopted from a recently developed signal enhancement technique, is aimed at recovering the noise-free output signal form the noise-contaminated signal by exploiting the algebraic properties of the ideal data matrix. Numerical examples show that the proposed method behaves much better than some other popular methods, especially in colored noise