Open AccessPARAMETER ESTIMATION SPACE FOR UNKNOWN INTERNAL EVOLUTION ON IOT DOMOTIC SYSTEMS
Author(s) -
Ricardo Carreño Aguilera,
J. J. Medel Juárez,
Sandra L. Gomez-Coronel
Publication year - 2020
Publication title -
fractals
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.654
H-Index - 44
eISSN - 1793-6543
pISSN - 0218-348X
DOI - 10.1142/s0218348x20500668
Subject(s) - computer science , probabilistic logic , internet of things , operator (biology) , covariance , convergence (economics) , estimation theory , function (biology) , estimation , covariance operator , rate of convergence , parameter space , mathematical optimization , machine learning , artificial intelligence , algorithm , mathematics , statistics , key (lock) , biology , embedded system , biochemistry , management , evolutionary biology , transcription factor , economics , gene , computer security , repressor , economic growth , chemistry
This paper describes the parameter estimation modeling concerning a domotic designer bot system with internet of things (IoT) assistance using the probabilistic operator based on the stochastic parameter estimation through the moments and the recursive conditions. Light, CCTV, presence, and temperature are IoT data monitored, shared, and accessed by the internet for a smart office designer performance that evolves based on historical web data. The relationship established by Wiener between covariance and variance found the parameter time evolution by observing through the time. The development is viewed in the visible results between non-recursive and recursive mathematical structures. In both cases, the convergence rate is based on probabilistic estimation, the functional error presents a high convergence rate which is viewed as an effect of the function of a density function. The estimate considered a non-invasive perspective, and it helps in different applications such as health diagnosis in humans and animals with internal problems, or systems which are unknown for internal evolution such as for IoT model adoption. Therefore, our objective is to propose a black box, inner approximation through the parameter estimation without a no invasive stochastic method based in Wiener approximation.