Title : 
2-D-Recursive Modelling of Homogeneous Discrete Gaussian Markov Fields
         
        
            Author : 
Carravetta, Francesco
         
        
            Author_Institution : 
Ist. di Analisi dei Sist. ed Inf. Antonio Ruberti, Consiglio Naz. delle Ric., Rome, Italy
         
        
        
        
        
            fDate : 
5/1/2011 12:00:00 AM
         
        
        
        
            Abstract : 
A 2-D-stochastic realization is obtained for an homogeneous discrete Gaussian Markov-field in two dimensions. In specific, it is shown that such Markov fields can be represented as the (unique) pathwise solution of a 2-D-recursive system of equations, with a locally-correlated “forcing-noise.” The idea of stochastic “dynamic” system in two dimensions gets realized this way. Two types of domain are considered: spheric and rectangular.
         
        
            Keywords : 
Gaussian processes; Markov processes; discrete systems; stochastic systems; 2D recursive modelling; homogeneous discrete Gaussian Markov field; stochastic dynamic system; Correlation; Equations; Kernel; Markov processes; Mathematical model; Nickel; Pixel; Modeling; realization theory; stochastic fields; stochastic systems;
         
        
        
            Journal_Title : 
Automatic Control, IEEE Transactions on
         
        
        
        
        
            DOI : 
10.1109/TAC.2011.2107114