ROBUST ESTIMATION FOR NOISY DATA

We developed a novel technique of robust estimation of the discrete Hilbert transform (DHT) of noisy geophysical data. The technique used the sinc method, in which the data were transformed via conformal mapping and the sinc bases were determined by solving a linear matrix equation. A transformation rule was presented for selecting a suitable conformal mapping function that would transform the class of geophysical data set in an appropriate interval range. A novel regularization technique was designed to obtain a robust solution of sinc bases when the data contained noise, in which an optimal regularization parameter was obtained in an automated way using a 1D optimization scheme. The technique of selecting the optimal value of the regularization parameter required no a priori knowledge about the level of noise contamination in the data. Numerical experiments were conducted on synthetically generated and published field data sets with a varying level of noise contamination to test the performance of the scheme. The results obtained using the proposed technique of DHT and those obtained by a standard Fourier domain technique were compared, and it was established that the proposed scheme of discrete Hilbert transformation performed better than that of the standard Fourier domain technique, for noise free and noisy data. The scheme was applied successfully on potential field and infrasound waveform data and also in estimating instantaneous frequency of nonstationary ultrasonic waveform data, which suggested applicability of the scheme to a wide class of geophysical data.

Download Full-text

Robust estimation of camera location and orientation from noisy data having outliers

[1989] Proceedings. Workshop on Interpretation of 3D Scenes ◽

10.1109/tdscen.1989.68101 ◽

2003 ◽

Cited By ~ 19

Author(s):

R. Kumar ◽

A.R. Hanson

Keyword(s):

Robust Estimation ◽

Noisy Data ◽

Camera Location

Download Full-text

Modified maximum likelihood method for the robust estimation of system parameters from very noisy data

Automatica ◽

10.1016/0005-1098(86)90085-3 ◽

1986 ◽

Vol 22 (2) ◽

pp. 231-235 ◽

Cited By ~ 58

Author(s):

S. Puthenpura ◽

N.K. Sinha

Keyword(s):

Maximum Likelihood ◽

Robust Estimation ◽

Maximum Likelihood Method ◽

Noisy Data ◽

Likelihood Method ◽

System Parameters ◽

Modified Maximum Likelihood

Download Full-text

SPE Approach for Robust Estimation of SIR Model with Limited and Noisy Data: The Case for COVID-19

Disaster Medicine and Public Health Preparedness ◽

10.1017/dmp.2020.220 ◽

2020 ◽

pp. 1-22

Author(s):

Kerem Senel ◽

Mesut Ozdinc ◽

Selcen Ozturkcan

Keyword(s):

Robust Estimation ◽

Potential Problem ◽

Noisy Data ◽

Sir Model ◽

Decision Makers ◽

Parameter Estimates ◽

New Approach ◽

Significant Challenge ◽

Research Questions ◽

Model Variation

Abstract The SIR model and its variants are widely used to predict the progress of COVID-19 worldwide, despite their rather simplistic nature. Nevertheless, robust estimation of the SIR model presents a significant challenge, particularly with limited and possibly noisy data in the initial phase of the pandemic. K-means algorithm is used to perform a cluster analysis of the top ten countries with the highest number of COVID-19 cases, to observe if there are any significant differences among countries in terms of robustness. As a result of model variation tests, the robustness of parameter estimates is found to be particularly problematic in developing countries. The incompatibility of parameter estimates with the observed characteristics of COVID-19 is another potential problem. Hence, a series of research questions are visited. We propose a SPE (“Single Parameter Estimation”) approach to circumvent these potential problems if the basic SIR is the model of choice, and we check the robustness of this new approach by model variation and structured permutation tests. Dissemination of quality predictions is critical for policy and decision-makers in shedding light on the next phases of the pandemic.

Download Full-text