Estudio e implementación de algoritmos de inferencia Bayesiana en sistemas espacio-temporales

e-Archivo Repository

Show simple item record

dc.contributor.advisor Elvira Arregui, Víctor Martín Gutiérrez, David 2018-06-26T10:22:34Z 2018-06-26T10:22:34Z 2016-10 2016-10-03
dc.description.abstract Many problems in engineering require estimation of the state of a system which changes over the time using a set of noisy measurements made on the system. In this project we focus on the state-space approach to modelling dynamic systems. First of all we study the Kalman filter algorithm which achieves the optimal solution in linear and Gaussian models. The Kalman filter minimises the variance of the estimation error. In nonlinear and/or Non-Gaussian models, approximations to the distribution of interest must be performed. We study some suboptimal algorithms such as the Monte Carlo methods and in particular, we focus on the Particle filter which is a Sequential Monte Carlo method. Over the project, several experiments in MATLAB are done with the goal of discussing and comparing the algorithms performances in several situations to demonstrate their theoretical features.
dc.format.mimetype application/pdf
dc.language.iso spa
dc.rights Atribución-NoComercial-SinDerivadas 3.0 España
dc.subject.other Kalman filter
dc.subject.other Particle filter
dc.subject.other Monte Carlo methods
dc.subject.other Bayesian inference
dc.subject.other State-Space models
dc.subject.other Sequential Importance Sampling
dc.title Estudio e implementación de algoritmos de inferencia Bayesiana en sistemas espacio-temporales
dc.type bachelorThesis
dc.subject.eciencia Telecomunicaciones
dc.rights.accessRights openAccess Ingeniería de Sistemas Audiovisuales
dc.contributor.departamento Universidad Carlos III de Madrid. Departamento de Teoría de la Señal y Comunicaciones
 Find Full text

Files in this item

*Click on file's image for preview. (Embargoed files's preview is not supported)

The following license files are associated with this item:

This item appears in the following Collection(s)

Show simple item record