Showing papers by "John B. Moore published in 1993"
TL;DR: The online EM schemes have significantly reduced memory requirements and improved convergence, and they can estimate HMM parameters that vary slowly with time or undergo infrequent jump changes.
Abstract: Sequential or online hidden Markov model (HMM) signal processing schemes are derived, and their performance is illustrated by simulation. The online algorithms are sequential expectation maximization (EM) schemes and are derived by using stochastic approximations to maximize the Kullback-Leibler information measure. The schemes can be implemented either as filters or fixed-lag or sawtooth-lag smoothers. They yield estimates of the HMM parameters including transition probabilities, Markov state levels, and noise variance. In contrast to the offline EM algorithm (Baum-Welch scheme), which uses the fixed-interval forward-backward scheme, the online schemes have significantly reduced memory requirements and improved convergence, and they can estimate HMM parameters that vary slowly with time or undergo infrequent jump changes. Similar techniques are used to derive online schemes for extracting finite-state Markov chains imbedded in a mixture of white Gaussian noise (WGN) and deterministic signals of known functional form with unknown parameters. >
289 citations
TL;DR: Expectation maximization algorithms are used to extract discrete-time finite-state Markov signals imbedded in a mixture of Gaussian white-noise and deterministic signals of known functional form with unknown parameters.
Abstract: Expectation maximization algorithms are used to extract discrete-time finite-state Markov signals imbedded in a mixture of Gaussian white-noise and deterministic signals of known functional form with unknown parameters. Maximum-likelihood estimates of the Markov state levels, state estimates, transition possibilities, and the parameters of the deterministic signals are obtained. Two types of deterministic signals are considered: periodic, or almost periodic signals with unknown frequency components, amplitudes, and phases; and polynomial drift in the states of the Markov process with the coefficients of the polynomial unknown. The techniques and supporting theory appear more elegant and powerful than ad hoc heuristic alternatives. An illustrative application to extracting ionic channel currents in cell membranes in the presence of white Gaussian noise and AC hum is included. >
27 citations
02 Jun 1993
TL;DR: This paper considers the linear quadratic problem with static output feedback and shows that an optimal solution can be successfully computed by finding the limiting solution of an ordinary differential equation given in terms of the gradient flow associated with the cost function.
Abstract: This paper considers the linear quadratic problem with static output feedback. It is shown that an optimal solution can be successfully computed by finding the limiting solution of an ordinary differential equation which is given in terms of the gradient flow associated with the cost function. Several properties are obtained concerning the gradient flow. For example, it is shown that the flow contains a subsequence convergent to a locally optimal output feedback gain. In the special case of state feedback, the flow is guaranteed to converge to the optimal gain. The effectiveness of the method is demonstrated by an example.
16 citations
15 Dec 1993
TL;DR: The proposed algorithm provides an important tool in investigating the complex structure of the pole placement task as well as a means to compute optimal output feedback gain matrices to provide a "best fit" between the system poles and the desired poles.
Abstract: In this paper we propose a numerical algorithm for determining optimal output feedback gains for the pole placement task for symmetric state space systems. The algorithm is based on minimising a least squares cost criterion which is well defined even when an exact solution to the pole placement task does not exist. Thus, the proposed algorithm provides an important tool in investigating the complex structure of the pole placement task as well as a means to compute optimal output feedback gain matrices to provide a "best fit" between the system poles and the desired poles. >
10 citations
TL;DR: In contrast to the off-line Expectation Maximisation (EM) algorithm, the on-line schemes have significantly reduced memory requirements, and with appropriate forgetting, can track slowly varying HMM parameters in an asymptotically efficient manner.
7 citations
TL;DR: In this paper, a class of nonlinear regulator optimal control problems with an infinite planning horizon is considered, where the original problem is reduced to a constrained optimal parameter selection problem and the question concerning the asymptotical stability of the dynamical system under the obtained optimal feedback controller is investigated.
5 citations
Proceedings Article•
01 Oct 1993
TL;DR: In this paper, a maximum likelihood estimation scheme for finite-state semi-Markov chains in white Gaussian noise is proposed. But the authors assume that the transition probabilities of known parametric from with unknown parameters.
Abstract: This paper develops maximum likelihood (ML) estimation schemes for finite-state semi-Markov chains in white Gaussian noise. We assume that the semi-Markov chain is characterised by transition probabilities of known parametric from with unknown parameters. We reformulate this hidden semi-Markov model (HSM) problem in the scalar case as a two-vector homogeneous hidden Markov model (HMM) problem in which the state consist of the signal augmented by the time to last transition. With this reformulation we apply the expectation Maximumisation (EM ) algorithm to obtain ML estimates of the transition probabilities parameters, Markov state levels and noise variance. To demonstrate our proposed schemes, motivated by neuro-biological applications, we use a damped sinusoidal parameterised function for the transition probabilities.
3 citations
TL;DR: In this article, a discrete time, partially observed control problem is discussed by explicitly constructing a reference probability in which the observations are independent, and using the unnormalized conditional probabilities as information states the problem is treated in separated form.
2 citations
01 Nov 1993
TL;DR: In this article, a recursive equation for the unnormalized joint conditional density of a noisily observed Markov chain and parameters which determine the transition densities and coefficients in the observations are obtained.
Abstract: Using the reference probability method, a recursive equation is obtained for the unnormalized joint conditional density of a noisily observed Markov chain, and parameters which determine the transition densities and coefficients in the observations. >
2 citations
TL;DR: The proposed method of control is motivated toward the control of flexible structures, such as large space structures, exhibiting large model order, model uncertainty, and decentralisation.
1 citations