An extended RLS type algorithm based on a non-linear function of the error
DOI: 163 Downloads 5162 Views
Author(s)
Abstract
Over the last few years, extended recursive and kernelized algorithms were one of the most promising in terms of tracking signals of state-space models in non-stationary environments. In this work, we intend to propose an EX-RLS (Extended Recursive Least Squares) algorithm based on a non-linear sum function of the error. The simulations were made in the problem by tracking a non-linear Rayleigh fading multipath channel. The results showed that the proposed algorithm exhibits a superior signal tracking capability than the kernelized extended recursive type versions.
Keywords
Recursive filter adaptive, ex-rnl algorithm, nonquadratic function, tracking, convergence rate
Cite this paper
L.F. Coelho Amaral, M. Vinicius Lopes, A. K. Barros,
An extended RLS type algorithm based on a non-linear function of the error
, SCIREA Journal of Electrical Engineering.
Volume 5, Issue 6, December 2020 | PP. 136-140.
References
[ 1 ] | A. H. Sayed. Fundamentals of Adaptive Filtering, HOBOKEN, NJ, USA: Wiley, 2003. |
[ 2 ] | L. F. C. Amaral, M. V. Lopes, A. K. Barros. A nonquadratic algorithm based on the extended recursive least squares algorithm, IEEE SIGNAL PROCESSING LETTERS, vol. 25, n0. 10, October 2018.5, Issue 10, 2018, pp. 1535-1539. |
[ 3 ] | W. Liu, I. Park, Y. Wang, and J. C. Príncipe. Extended kernel recursive least squares algorithm, IEEE Trans. Signal Process., vol 57, n0 10, pp. 3801-3814. |
[ 4 ] | E. Walach and B. Widdrow. The least mean fourth (LMF) adaptive algorithm and its family, IEEE Trans. Inf. Theory, vol. IT-30, n0 2, pp. 275-283. |