In this paper, a factor graph approach is employed to investigate the recursive filtering problem for conditionally linear Gaussian state-space models. First, we derive a new factor graph for the considered filtering problem; then, we show that applying the sum-product rule to our graphical model results in both known and novel filtering techniques. In particular, we prove that: 1) marginalized particle filtering can be interpreted as a form of forward only message passing over the devised graph; 2) novel filtering methods can be easily developed by exploiting the graph structure and/or simplifying probabilistic messages.
Marginalized Particle Filtering and Related Filtering Techniques as Message Passing / Vitetta, Giorgio M.; Sirignano, Emilio; Di Viesti, Pasquale; Montorsi, Francesco; Sola, Matteo. - In: IEEE TRANSACTIONS ON SIGNAL PROCESSING. - ISSN 1053-587X. - ELETTRONICO. - 67:6(2019), pp. 1522-1536. [10.1109/TSP.2019.2893868]
Marginalized Particle Filtering and Related Filtering Techniques as Message Passing
Di Viesti, Pasquale;
2019
Abstract
In this paper, a factor graph approach is employed to investigate the recursive filtering problem for conditionally linear Gaussian state-space models. First, we derive a new factor graph for the considered filtering problem; then, we show that applying the sum-product rule to our graphical model results in both known and novel filtering techniques. In particular, we prove that: 1) marginalized particle filtering can be interpreted as a form of forward only message passing over the devised graph; 2) novel filtering methods can be easily developed by exploiting the graph structure and/or simplifying probabilistic messages.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.