The reconstruction of Gaussian processes realizations with an arbitrary set of samples

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

The method of the reconstruction of Gaussian processes realizations based on the Conditional Mean Rule is described. The optimal reconstruction algorithms of stationary and non stationary processes are considered. In result of the investigation, two principal characteristics (the reconstruction function and the error reconstruction function) are obtained for all variants of processes. There are two features of studied algorithms: the reconstruction function is a linear function of samples and the error reconstruction function does not depend on the values of samples. These features are valid for the sampling- reconstruction procedure of Gaussian processes only.

Original languageEnglish
Title of host publicationRecent Researches in Telecommunications, Informatics, Electronics and Signal Processing - TELE-INFO'11, MINO'11, SIP'11
Pages120-125
Number of pages6
StatePublished - 2011
Event10th WSEAS International Conf. on Telecommunications and Informatics, TELE-INFO'11, 10th WSEAS International Conference on Microelectronics, Nanoelectronics, Optoelectronics, MINO'11, 10th WSEAS International Conference on Signal Processing, SIP'11 - Lanzarote, Canary Islands, Spain
Duration: 27 May 201129 May 2011

Publication series

NameRecent Researches in Telecommunications, Informatics, Electronics and Signal Processing - TELE-INFO'11, MINO'11, SIP'11

Conference

Conference10th WSEAS International Conf. on Telecommunications and Informatics, TELE-INFO'11, 10th WSEAS International Conference on Microelectronics, Nanoelectronics, Optoelectronics, MINO'11, 10th WSEAS International Conference on Signal Processing, SIP'11
Country/TerritorySpain
CityLanzarote, Canary Islands
Period27/05/1129/05/11

Keywords

  • Error reconstruction
  • Gaussian process
  • Sampling-reconstruction algorithm

Fingerprint

Dive into the research topics of 'The reconstruction of Gaussian processes realizations with an arbitrary set of samples'. Together they form a unique fingerprint.

Cite this