A comparison of least squares and maximum likelihood methods using sine fitting in ADC testing

Ján Šaliga, I. Kollár, Linus Michaeli, Ján Buša, Jozef Lipták, Tamás Virosztek

Research output: Contribution to journalArticle

25 Citations (Scopus)

Abstract

ADC test methods require the best possible reconstruction of the input signal of the ADC under test from the acquired, therefore erroneous, ADC output data. The commonly used least squares (LS) fit and the recently introduced maximum likelihood (ML) estimation are competing methods. This paper presents a simulation-based comparative study of these estimation methods with the goal to investigate the behavior of both methods and to determine their limits. Two alternative algorithms for the calculation of the maximum likelihood fit are considered (gradient-based minimization and differential evolution). The main finding is that while for low-INL (linear) ADCs the two methods (LS and ML) give similar results, for practical (almost always nonlinear) ADCs ML is definitely better.

Original languageEnglish
Pages (from-to)4362-4368
Number of pages7
JournalMeasurement: Journal of the International Measurement Confederation
Volume46
Issue number10
DOIs
Publication statusPublished - 2013

Fingerprint

Maximum Likelihood Method
Maximum likelihood
Least Squares
Maximum Likelihood
Testing
Maximum likelihood estimation
least squares method
Differential Evolution
Least Square Method
Maximum Likelihood Estimation
Comparative Study
Gradient
gradients
optimization
output
Output
Alternatives
Simulation
simulation

Keywords

  • ADC test
  • Differential evolution
  • Estimation of signal parameters
  • Four-parameter fit
  • Gradient-based method
  • Least squares method
  • LS method
  • Maximum likelihood estimation
  • Signal recovery

ASJC Scopus subject areas

  • Condensed Matter Physics
  • Applied Mathematics

Cite this

A comparison of least squares and maximum likelihood methods using sine fitting in ADC testing. / Šaliga, Ján; Kollár, I.; Michaeli, Linus; Buša, Ján; Lipták, Jozef; Virosztek, Tamás.

In: Measurement: Journal of the International Measurement Confederation, Vol. 46, No. 10, 2013, p. 4362-4368.

Research output: Contribution to journalArticle

Šaliga, Ján ; Kollár, I. ; Michaeli, Linus ; Buša, Ján ; Lipták, Jozef ; Virosztek, Tamás. / A comparison of least squares and maximum likelihood methods using sine fitting in ADC testing. In: Measurement: Journal of the International Measurement Confederation. 2013 ; Vol. 46, No. 10. pp. 4362-4368.
@article{8c09cea552d5436794705d231b367d44,
title = "A comparison of least squares and maximum likelihood methods using sine fitting in ADC testing",
abstract = "ADC test methods require the best possible reconstruction of the input signal of the ADC under test from the acquired, therefore erroneous, ADC output data. The commonly used least squares (LS) fit and the recently introduced maximum likelihood (ML) estimation are competing methods. This paper presents a simulation-based comparative study of these estimation methods with the goal to investigate the behavior of both methods and to determine their limits. Two alternative algorithms for the calculation of the maximum likelihood fit are considered (gradient-based minimization and differential evolution). The main finding is that while for low-INL (linear) ADCs the two methods (LS and ML) give similar results, for practical (almost always nonlinear) ADCs ML is definitely better.",
keywords = "ADC test, Differential evolution, Estimation of signal parameters, Four-parameter fit, Gradient-based method, Least squares method, LS method, Maximum likelihood estimation, Signal recovery",
author = "J{\'a}n Šaliga and I. Koll{\'a}r and Linus Michaeli and J{\'a}n Buša and Jozef Lipt{\'a}k and Tam{\'a}s Virosztek",
year = "2013",
doi = "10.1016/j.measurement.2013.05.004",
language = "English",
volume = "46",
pages = "4362--4368",
journal = "Measurement: Journal of the International Measurement Confederation",
issn = "1536-6367",
publisher = "Elsevier",
number = "10",

}

TY - JOUR

T1 - A comparison of least squares and maximum likelihood methods using sine fitting in ADC testing

AU - Šaliga, Ján

AU - Kollár, I.

AU - Michaeli, Linus

AU - Buša, Ján

AU - Lipták, Jozef

AU - Virosztek, Tamás

PY - 2013

Y1 - 2013

N2 - ADC test methods require the best possible reconstruction of the input signal of the ADC under test from the acquired, therefore erroneous, ADC output data. The commonly used least squares (LS) fit and the recently introduced maximum likelihood (ML) estimation are competing methods. This paper presents a simulation-based comparative study of these estimation methods with the goal to investigate the behavior of both methods and to determine their limits. Two alternative algorithms for the calculation of the maximum likelihood fit are considered (gradient-based minimization and differential evolution). The main finding is that while for low-INL (linear) ADCs the two methods (LS and ML) give similar results, for practical (almost always nonlinear) ADCs ML is definitely better.

AB - ADC test methods require the best possible reconstruction of the input signal of the ADC under test from the acquired, therefore erroneous, ADC output data. The commonly used least squares (LS) fit and the recently introduced maximum likelihood (ML) estimation are competing methods. This paper presents a simulation-based comparative study of these estimation methods with the goal to investigate the behavior of both methods and to determine their limits. Two alternative algorithms for the calculation of the maximum likelihood fit are considered (gradient-based minimization and differential evolution). The main finding is that while for low-INL (linear) ADCs the two methods (LS and ML) give similar results, for practical (almost always nonlinear) ADCs ML is definitely better.

KW - ADC test

KW - Differential evolution

KW - Estimation of signal parameters

KW - Four-parameter fit

KW - Gradient-based method

KW - Least squares method

KW - LS method

KW - Maximum likelihood estimation

KW - Signal recovery

UR - http://www.scopus.com/inward/record.url?scp=84887045422&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84887045422&partnerID=8YFLogxK

U2 - 10.1016/j.measurement.2013.05.004

DO - 10.1016/j.measurement.2013.05.004

M3 - Article

AN - SCOPUS:84887045422

VL - 46

SP - 4362

EP - 4368

JO - Measurement: Journal of the International Measurement Confederation

JF - Measurement: Journal of the International Measurement Confederation

SN - 1536-6367

IS - 10

ER -