Title :
A split quaternion nonlinear adaptive filter
Author :
Ujang, Che Ahmad Bukhari Che ; Took, Clive Cheong ; Kavcic, Alek ; Mandic, Danilo P.
Author_Institution :
Electr. & Electron. Eng. Dept., Imperial Coll. London, London
Abstract :
A split quaternion learning algorithm for the training of nonlinear finite impulse response filters for the modelling of hypercomplex signals is proposed. A rigorous derivation takes into account the non-commutativity of the quaternion product, an aspect not taken into account in the existing nonlinear architectures, such as the quaternion multilayer perceptron (QMLP). It is shown that the additional information present within the proposed algorithm provides an improved performance over QMLP. Simulation on both benchmark and real-world signals support the approach.
Keywords :
FIR filters; adaptive filters; multilayer perceptrons; nonlinear filters; finite impulse response filters; hypercomplex signal modelling; quaternion multilayer perceptron; split quaternion learning algorithm; split quaternion nonlinear adaptive filter; Adaptive filters; Finite impulse response filter; Machine learning; Machine learning algorithms; Multidimensional systems; Multilayer perceptrons; Neural networks; Nonlinear equations; Quaternions; Signal processing; Cauchy-Riemann-Fueter equation; Machine Learning; Quaternion Multilayer Perceptron;
Conference_Titel :
Acoustics, Speech and Signal Processing, 2009. ICASSP 2009. IEEE International Conference on
Conference_Location :
Taipei
Print_ISBN :
978-1-4244-2353-8
Electronic_ISBN :
1520-6149
DOI :
10.1109/ICASSP.2009.4959941