Immune-Inspired Optimization with Autocorrentropy Function for Blind Inversion of Wiener Systems

Author(s)

Fernandez, Stephanie A. ; Fantinato, Denis G. ; Montalvao, Jugurta ; Attux, Romis ; Silva, Daniel G. ; IEEE

• Document type: Journal article
• Source: 2018 IEEE CONGRESS ON EVOLUTIONARY COMPUTATION (CEC); v. N/A, p. 7-pg., 2018-01-01.

Abstract

Blind inversion of nonlinear systems is a complex task that requires some sort of prior information about the source e.g. whether it is composed of independent samples or, particularly in this work, a dependence "signature" which is assumed to be known via the autocorrentropy function. Furthermore, it involves the solution of a nonlinear, multimodal optimization problem to determine the parameters of the inverse model. Thus, we propose a blind method for Wiener systems inversion, which is composed of a correntropy-based criterion in association to the well-known CLONALG immune-inspired optimization metaheuristic. The empirical results validate the methodology for continuous and discrete signals. (AU)

• FAPESP's process: 17/11488-5 - Multivariate Analysis of the Data Temporal Structure for Blind Source Separation in the Context of Nonlinear Mixtures and of Multiple Datasets
• Grantee: Denis Gustavo Fantinato
• Support Opportunities: Scholarships in Brazil - Post-Doctoral