6533b7d7fe1ef96bd1268e2c

RESEARCH PRODUCT

Non-linear RLS-based algorithm for pattern classification

Antonio J. Serrano-lópezJavier Calpe-maravillaGustau Camps-vallsJosé D. Martín-guerreroEmilio Soria-olivasJoan Vila-francés

subject

Recursive least squares filterSignal processingEqualizationContext (language use)Filter (signal processing)Computer Science::OtherNonlinear systemComputer Science::SoundControl and Systems EngineeringSignal ProcessingComputer Vision and Pattern RecognitionElectrical and Electronic EngineeringAlgorithmSoftwareMathematics

description

A new non-linear recursive least squares (RLS) algorithm is presented in the context of pattern classification problems. The algorithm incorporates the non-linearity of the filter's output in the updating rules of the classical RLS algorithm. The proposed method yields lower stationary error levels when compared to the standard LMS and RLS algorithms in a classical application of pattern classification, such as the channel equalization problem.

yearjournalcountryeditionlanguage
2006-05-01Signal Processing
https://doi.org/10.1016/j.sigpro.2005.09.004