Subspace system identification considering both noise attenuation and use of prior knowledge

This paper proposes a new subspace state-space system identification method which consists of two stages. At the first stage, the noise attenuation is achieved based on the uncorrelation between input signal and noises, where a large amount of data can be handled with the prescribed size matrices. At the second stage, the available prior knowledge on the plant is fully used for the subspace based identification. The combination of these two methods enable us to obtain accurate state-space models. An illustrative numerical example is given to show the effectiveness of the proposed method.