Combinations of adaptive filters and techniques of machine learning

• Grant number: 11/13581-6
• Support Opportunities: Scholarships abroad - Research
• Start date: January 16, 2012
• End date: July 15, 2012
• Field of knowledge: Engineering - Electrical Engineering - Telecommunications
• Principal Investigator: Magno Teófilo Madeira da Silva
• Grantee: Magno Teófilo Madeira da Silva
• Host Investigator: Jerónimo Arenas-García
• Host Institution: Escola Politécnica (EP). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• Institution abroad: Universidad Carlos III de Madrid (UC3M), Spain

Abstract

Combinations of filters operating in parallel have attracted attention in the signal processing community. These combinations have been proposed to improve the performance of adaptive filters since they present universal behavior, i.e., considering stationary inputs, the combined estimate is at least as good as the best component filter. Therefore, they have been used to improve the choice of fixed parameters of the filters, tracking capability, stability, steady-state solution, initialization, among other factors. Although many results related to this subject have been published, there are some open issues, which will be be addressed in this research. Among them, we can mention: (i) the transient analysis of the convex combination, (ii) comparison among convex, affine, and linear combinations, taking into account different simulation scenarios, (iii) combination of the constant modulus algorithm (CMA) with the LMS (least mean-squares) algorithm in the directed-decision scheme to improve the performance of the blind equalizer, and (iv) the improvement of the performance of combinations by imposing diversity to reduce the cross-correlation between the outputs of the filter components. We also intend to study machine learning techniques to accelerate the convergence of blind algorithms for equalization of high-order QAM (quadrature amplitude modulation) signals. (AU)

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

SILVA, MAGNO T. M.; ARENAS-GARCIA, JERONIMO. A Soft-Switching Blind Equalization Scheme via Convex Combination of Adaptive Filters. IEEE TRANSACTIONS ON SIGNAL PROCESSING, v. 61, n. 5, p. 1171-1182, MAR 2013. (11/13581-6)