Maximum likelihood estimation in the context of an optical measurement - Archive ouverte HAL Accéder directement au contenu
Chapitre D'ouvrage Année : 2020

Maximum likelihood estimation in the context of an optical measurement

Miguel Alonso

Résumé

Maximum likelihood estimation (MLE) is a widely-used statistical approach for estimating one or more unknown parameters of a probabilistic model based on observed data. In this tutorial, we review MLE's mathematical foundations within the context of measurements of optical intensity distributions. Here, the detection of each photon is treated as a random event, the outcome being the photon's location within a pixelized array. These detected photons accumulate to form an intensity profile. We give a straightforward derivation for the likelihood function and Fisher information matrix (FIM) associated with a measurement. An estimate for the parameter(s) of interest is then obtained by maximizing the likelihood function, while the FIM determines the estimate's uncertainty. These concepts are illustrated with several simple examples involving a small number of pixels, for one and two parameters, revealing interesting properties and practical considerations for optical measurements. Connections are drawn to weak (off-null) measurements.
Fichier principal
Vignette du fichier
MLE_optics_tutorial_v4_8.pdf (19.88 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02554686 , version 1 (01-06-2020)

Identifiants

Citer

Anthony Vella, Miguel Alonso. Maximum likelihood estimation in the context of an optical measurement. A Tribute to Emil Wolf, 65, pp.231-311, 2020, Progress in optics, ⟨10.1016/bs.po.2019.11.007⟩. ⟨hal-02554686⟩
57 Consultations
215 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More