Recursive and iterative stochastic gradient algorithms based on two-step update estimation for wiener systems

Abstract

In this paper, a recursive stochastic gradient algorithm based on two-step update estimation and an iterative stochastic gradient algorithm based on two-step update estimation are established for the Wiener system by introducing a relaxation factor, which controls the relative importance of the two estimation parts. In addition, the convergence performance of the proposed SG-TSU and ISG-TSU algorithms are then analyzed. It is shown by a numerical example that if the relaxation factor is appropriately chosen, the proposed SG-TSU and ISG-TSU algorithms converge more quickly and have higher convergence precision than the standard SG and ISG algorithms, respectively.