Identification of jump Markov linear models using particle filters
This work addresses a specific modeling challenge in signal processing or control systems, representing an incremental improvement through algorithmic innovation.
The authors tackled the challenging problem of identifying jump Markov linear models, which lack an analytical solution, by deriving a new expectation maximization algorithm that produces maximum likelihood estimates of model parameters.
Jump Markov linear models consists of a finite number of linear state space models and a discrete variable encoding the jumps (or switches) between the different linear models. Identifying jump Markov linear models makes for a challenging problem lacking an analytical solution. We derive a new expectation maximization (EM) type algorithm that produce maximum likelihood estimates of the model parameters. Our development hinges upon recent progress in combining particle filters with Markov chain Monte Carlo methods in solving the nonlinear state smoothing problem inherent in the EM formulation. Key to our development is that we exploit a conditionally linear Gaussian substructure in the model, allowing for an efficient algorithm.