Stochastic gradient based iterative identification algorithm for a class of dual-rate Wiener systems

Parameter estimation problem is considered for a class of dual-rate Wiener systems whose input-output data are measured by two different sampling rate. Firstly, a polynomial transformation technique is used to derive a mathematical model for such dual-rate Wiener systems. Then, directly based on the dual-rate sampled data, a dual-rate Wiener systems stochastic gradient algorithm (DRW-SG) is presented. In order to improve the algorithm convergence rate, a dual-rate Wiener systems stochastic gradient algorithm with a forgetting factor algorithm (DRW-FF-SG) is presented. For making full use of the forgetting factor, a dual-rate Wiener systems stochastic gradient algorithm with an increasing forgetting factor algorithm (DRW-IFF-SG) is presented which performs excellently. Finally, an example is provided to test and illustrate the proposed algorithms.