Technion You are invited to attend a lecture by Prof. Yarim Ephraim Department of ECE, George Mason University Title: Parameter Estimation for Bivariate Markov Chains Date: Wednesday, 6/7/2011 Time: at 13:30 Place: in room 1066 Electrical Eng. Building Technion City The original announcement for this is now at <http://www.math.technion.ac.il/~techm/temp/20110706Biva.pdf> It will be available for up to a month after the date of the activity. Abstract: Bivariate Markov chains are used in many applications such as communication network modeling, ion-channel currents estimation, and phylogenetics analysis. Only one process of the bivariate Markov chain is observable, while the other plays the role of an underlying process. Bivariate Markov chains are Markov renewal processes, and they may also be seen as hidden Markov processes. Some well-known examples of bivariate Markov chains include the Markov modulated Poisson process and the batch Markovian arrival process. In these examples, the generator of the bivariate Markov chain is such that the underlying process is a Markov chain. Moreover, in the first example, the underlying and observable processes cannot jump simultaneously. In this talk, we present some of our recent results on signal and parameter estimation of a finite-state homogeneous general bivariate Markov chain for which the underlying process need not be Markov and the two processes may jump simultaneously. We have developed explicit recursions for some of these problems which do not require any transformation or sampling scheme of the observed signal. . This is joint work with Brian L. Mark. . The work was supported by the U.S. National Science Foundation under Grant 0916568. --------------------------------------------------------- Technion Math. Net (TECHMATH) Editor: Michael Cwikel <techm@math.technion.ac.il> To see today's activities and other future and past activities go to <http://www.math.technion.ac.il/~techm/today.html> Announcement from: Rami Atar <atar@ee.technion.ac.il> ----------