The widely linear quaternion recursive total least squares
Thiannithi Thanthawaritthisai, Felipe A. Tobar, Anthony G. Constantinides, Danilo P. Mandic
Abstract
A widely linear quaternion recursive total least squares (WL-QRTLS) algorithm is introduced for the processing of ℚ-improper processes contaminated by noise. The total least squares for quaternions (QTLS) is a generalisation of the real-valued total least squares and is introduced rigorously, starting from the existence condition for low-rank approximation of quaternion matrices. Then, a quaternion Rayleigh quotient (QRQ) is defined to establish the link between the QTLS solution and the minimisation of the QRQ. Finally, the rank-one update formula is employed to allow for fast iterative solution based on the QRQ. Through simulations, the WL-QRTLS was shown to exhibit superior performance, under perturbations on both input and output signals, to other adaptive filtering of the same class - the widely linear quaternion least mean squares (WL-QLMS) and the widely linear quaternion recursive least squares (WL-QRLS). The experiments on both synthetic and real-world ℚ-improper processes supported the analysis.
BibTeX
@inproceedings{icassp2015_thewidelylinearq,
title = {The widely linear quaternion recursive total least squares},
author = {Thiannithi Thanthawaritthisai and Felipe A. Tobar and Anthony G. Constantinides and Danilo P. Mandic},
booktitle = {ICASSP 2015},
year = {2015}
}