Analysis of the Correntropy-Based Criterion for Blind Equalization with Precoded Sources
Vinícius A. de Oliveira, Denis G. Fantinato, Rafael Ferrari, Romis Attux, Levy Boccato

DOI: 10.14209/sbrt.2017.191
Evento: XXXV Simpósio Brasileiro de Telecomunicações e Processamento de Sinais (SBrT2017)
Keywords: Adaptive filtering Blind channel equalization Information-theoretic learning Correntropy
Abstract
Temporally structured sources may arise in blind equalization as the result of coding schemes. In this context, a promising approach involves the use of correntropy, which is capable of exploring both the statistical information and the temporal structure of the signals. In this work, we perform a detailed analysis of the correntropy-based criterion for equalization, giving a special attention to the effect of the adjustable parameters as well as to the comparison with the analytical correntropy measure, whose formal derivation is another contribution of this work. The experimental results indicate the attainable performance and the influence of the main parameters.

Download