Adaptive identification algorithm for continuous-time linear time-invariant systems in frequency domain

This paper considers the problem of identifying continuous-time linear time-invariant systems in frequency domain. An adaptive identification is proposed in terms of continuous rational orthogonal bases, where one-by-one selection of poles is performed for the basis functions. It is shown that the proposed algorithm achieves approximate optimality under these selections. The adaptivity of the proposed algorithm lies in that different poles are selected for basis functions for different systems. The effectiveness of the proposed adaptive algorithm is demonstrated by simulation results.