Improved system identification with Renormalization Group

- We present an improved system identification method.
- We create a new data set from the given data set based on Renormalization Group.
- We analyze both the asymptotic case and the finite-sample case.
- The proposed method is quite effective under certain conditions.

This paper proposes an improved system identification method with Renormalization Group. Renormalization Group is applied to a fine data set to obtain a coarse data set. The least squares algorithm is performed on the coarse data set. The theoretical analysis under certain conditions shows that the parameter estimation error could be reduced. The proposed method is illustrated with examples.