Impulsive noise detection in PLC with smoothed L0-norm

Authors

Filbert H. Juwono, University of Western Australia
Qinghua Guo, University of Wollongong, University of Western AustraliaFollow
Defeng (David) Huang, University of Western AustraliaFollow
Kitpo Wong, University of Western Australia
Lu Xu, University of Western AustraliaFollow

RIS ID

103921

Publication Details

F. H. Juwono, Q. Guo, D. Huang, K. Po. Wong & L. Xu, "Impulsive noise detection in PLC with smoothed L0-norm," in Acoustics, Speech and Signal Processing (ICASSP), 2015 IEEE International Conference on, 2015, pp. 3232-3236.

Abstract

Power-line communications (PLC) commonly employs orthogonal frequency-division multiplexing (OFDM) as the modulation technique, and impulsive noise has a significant negative impact on its performance. Using the property of null subcarriers in OFDM and the fact that impulsive noise is sparse, we formulate a minimization problem to detect and estimate the impulsive noise. In previous works, ℓ1-norm minimization was employed to achieve this task. In this paper, we propose to use a smoothed ℓ0-norm minimization algorithm for impulsive noise detection instead. Simulation results show that this approach is promising as it achieves comparable performance as ℓ1-norm minimization but with much lower complexity.