Bias Correction-Based Recursive Estimation for Dual-Rate Output-Error Systems with Sampling Noise

This paper develops a bias correction-based recursive estimation algorithm for dual-rate output-error systems. The system output is subjected to both output noise and sampling noise. Using the polynomial transformation technique, the dual-rate output-error system is converted into an identification model where the sampled data can be directly applied. The noise variances of output noise and sampling noise are estimated by solving a nonlinear equation, which can minimize the estimation errors of noise variances. The simulation examples demonstrate that the proposed algorithm has higher parameter estimation accuracy in contrast to the auxiliary model-based recursive least-squares algorithm.