Title :
On Ryde´n´s EM algorithm for estimating MMPPs
Author :
Roberts, William J.J. ; Ephraim, Yariv ; Dieguez, Elvis
Author_Institution :
Atlantic Coast Technol. Inc, Silver Spring, MD, USA
fDate :
6/1/2006 12:00:00 AM
Abstract :
Two aspects of Ryden´s expectation-maximization algorithm for estimating the parameter of a Markov modulated Poisson process are addressed. First, a scaling procedure is developed for the forward-backward recursions that circumvents the need for customized floating-point software. Second, evaluation of integrals of matrix exponentials is facilitated by applying a result due to Van Loan. For an MMPP of order four, a speedup of over two orders of magnitude was observed.
Keywords :
Markov processes; expectation-maximisation algorithm; matrix algebra; modulation; recursive estimation; MMPP; Markov modulated Poisson process; Ryden expectation-maximization algorithm; floating-point software; forward-backward recursions; matrix exponential; maximum likelihood estimation; parameter estimation; scaling procedure; Application software; Computer networks; Control theory; Expectation-maximization algorithms; Finance; Maximum likelihood estimation; Parameter estimation; Recursive estimation; Signal processing algorithms; State estimation; Expectation-maximization (EM) algorithm; Markov-modulated Poisson process; maximum likelihood estimation;
Journal_Title :
Signal Processing Letters, IEEE
DOI :
10.1109/LSP.2006.871709
