Aitken-based Acceleration Estimation Algorithms for a Nonlinear Model with Exponential Terms by Using the Decomposition

This paper studies some parameter estimation algorithms for a class of nonlinear models with exponential terms, i.e., the radial basis function-based state-dependent autoregressive (RBF-AR) models. An Aitken-based multi-innovation stochastic gradient algorithm is presented for the RBF-AR models based on the Aitken method. Inspired by the decomposition-coordination principle of large systems, an Aitken-based hierarchical multi-innovation stochastic gradient algorithm is proposed by combining the decomposition technique with the Aitken method. The effectiveness of the proposed algorithms are validated through two simulation examples.