학술논문

Diffusion Normalized Maximum Versoria Criterion Robust to Impulsive Noise
Document Type
Periodical
Author
Source
IEEE Transactions on Circuits and Systems II: Express Briefs IEEE Trans. Circuits Syst. II Circuits and Systems II: Express Briefs, IEEE Transactions on. 70(4):1660-1664 Apr, 2023
Subject
Components, Circuits, Devices and Systems
Steady-state
Convergence
Estimation
Robustness
Behavioral sciences
Noise measurement
Measurement uncertainty
Distributed estimation
diffusion-normalized maximum versoria criterion (d-NMVC)
impulsive noise
Language
ISSN
1549-7747
1558-3791
Abstract
In this brief, we propose a new diffusion normalized maximum Versoria criterion (d-NMVC) algorithm, which is based on maximization of the normalized maximum Versoria criterion (MVC) cost function to enhance the performance of the distributed estimation over networks in the presence of non-Gaussian noise. Convergence of the proposed algorithm, in the mean square sense and evolution behavior, under impulsive noise environment is also analyzed. Simulation results show the robustness of the proposed algorithm under impulsive noise environment against various non-Gaussian noise distributions.