Explicit filtering equations for labelled random finite sets

We decompose a probability density function (PDF) of a labelled random finite set (RFS) into a probability mass function over a set of labels and a PDF on a vector-valued multitarget state given the labels. Using this decomposition, we write the Bayesian filtering recursion for labelled RFSs in an e...

Full description

Bibliographic Details
Main Authors: Garcia Fernandez, Angel, Morelande, M.
Format: Conference Paper
Published: 2015
Online Access:http://hdl.handle.net/20.500.11937/4474
Description
Summary:We decompose a probability density function (PDF) of a labelled random finite set (RFS) into a probability mass function over a set of labels and a PDF on a vector-valued multitarget state given the labels. Using this decomposition, we write the Bayesian filtering recursion for labelled RFSs in an explicit form. The resulting formulas are of conceptual and practical interest in the RFS approach to multiple target tracking, especially, for track-before-detect particle filter implementations.