Comparing data scaling based recursive least squares algorithms with Kalman Filter for nano parameters identification

This paper considers a single-input and single-output (SISO) controlled autoregressive moving average system by using scalar factors of the input-output data. A general identification technique, through scaling data, is obtained. To obtain this data, Recursive Least Squares (RLS) methods are used to estimate the nano parameters of a linear model using input-output scaling factors. Different variations of the RLS method are tested and compared. The first RLS method uses a forgetting factor and the second method is integrated with a Kalman Filter covariance. In order to estimate the parameters in the nano range, the input signal requires a very high frequency and thus a very high sampling rate is required. Although using this proposed technique, a broader sampling rate and an input signal with low frequency can be used to identify the nano parameters characterizing the linear model. The simulation results indicate that the proposed algorithm is effective and robust. The main contribution of this work is to provide a scaled identification bandwidth and sampling rate of the detecting signal in the identification process.