English

Study of filtered-x logarithmic recursive least $p$-power algorithm

Signal Processing 2022-02-02 v1 Machine Learning

Abstract

For active impulsive noise control, a filtered-x recursive least pp-power (FxRLP) algorithm is proposed by minimizing the weighted summation of the pp-power of the \emph{a posteriori} errors. Since the characteristic of the target noise is investigated, the FxRLP algorithm achieves good performance and robustness. To obtain a better performance, we develop a filtered-x logarithmic recursive least pp-power (FxlogRLP) algorithm which integrates the pp-order moment with the logarithmic-order moment. Simulation results demonstrate that the FxlogRLP algorithm is superior to the existing algorithms in terms of convergence rate and noise reduction.

Keywords

Cite

@article{arxiv.2202.00560,
  title  = {Study of filtered-x logarithmic recursive least $p$-power algorithm},
  author = {Z. Zheng and L. Lu and Y. Yu and R. C. de Lamare and Z. Liu},
  journal= {arXiv preprint arXiv:2202.00560},
  year   = {2022}
}

Comments

7 pages, 3 figures

R2 v1 2026-06-24T09:13:48.928Z