F
Filipe Igreja
Researcher at Centro Federal de Educação Tecnológica Celso Suckow da Fonseca
Publications - 4
Citations - 33
Filipe Igreja is an academic researcher from Centro Federal de Educação Tecnológica Celso Suckow da Fonseca. The author has contributed to research in topics: Least mean squares filter & Independence (probability theory). The author has an hindex of 2, co-authored 3 publications receiving 17 citations.
Papers
More filters
Journal ArticleDOI
Exact Expectation Evaluation and Design of Variable Step-Size Adaptive Algorithms
TL;DR: It is argued that such a comparison can be misleading because the supposedly optimal step-size sequence sometimes induces divergence in the initial phase of learning, which occurs most often when the input signal is colored and/or heavy tailed.
Journal ArticleDOI
An Exact Expectation Model for the LMS Tracking Abilities
Thiago T. P. Silva,Pedro Lara,Filipe Igreja,Fernanda D. V. R. Oliveira,Luís Tarrataca,Diego B. Haddad +5 more
TL;DR: This work presents a comprehensive model of the performance of the least mean square algorithm, operating under Markovian time-varying channels, and is able to provide a deterministic theoretical step-size sequence that optimizes algorithmic performance, as well as an accurate step size upper bound that guarantees algorithm stability.
Journal ArticleDOI
On the Skewness of the LMS Adaptive Weights
TL;DR: This brief shows that the skewness of the adaptive weights distribution may present a large deviation from the common Gaussian assumption, especially in the first phase of the learning, and it is demonstrated that the skew may grow without limit even when adaptive weights present convergence in both average and mean square behaviors.
Journal ArticleDOI
Analysis of the least mean squares algorithm with reusing coefficient vector
Leonardo C. Resende,Rodrigo Pimenta,Newton N. Siqueira,Filipe Igreja,Diego B. Haddad,Mariane R. Petraglia +5 more
TL;DR: In this article , a deterministic and stochastic model is proposed to predict various learning characteristics of the LMS algorithm with coefficient reuse, which is able to improve the steady-state performance of adaptive filter algorithms, especially in very challenging low signal-to-noise scenarios.