Modeling continuous-time processes via input-to-state filters☆
Introduction
The identification of continuous-time stochastic processes is a fundamental research issue which has received considerable interest recently. Since many natural processes are continuous-time, it is of interest in many practical applications to identify a continuous-time model instead of a discrete-time model (Larsson, 2004). Although the signal is in continuous-time, in practice one works with sampled data. One popular approach is to identify a discrete-time system from uniformly sampled data, as shown in Larsson (2004) and references therein. Subsequently, the estimated discrete-time model is converted back to a continuous-time model via a nonlinear transformation (Söderström, 1991). This approach will be referred to as the indirect approach. Apart from the obvious difficulty of solving nonlinear equations, this approach also suffers from several other setbacks: (i) at a fast sampling rate, the poles and the zeros of the associated discrete-time system cluster close to the point in the complex plane, leading to a numerically ill-conditioned identification problem; (ii) the continuous-time parameters can be very sensitive to the sampled data. These issues have been discussed in detail in Fan, Söderström, Mossberg, Carlsson, and Zhou (1999), Fan, Söderström, and Zhou (1999), Larsson and Mossberg (2003), Larsson and Söderström (2002), Söderström (1991) and Söderström and Mossberg (2000). A second approach is to identify the continuous-time parameters directly. The basic idea here is to replace the differentiation operator with the delta operator (Feuer and Goodwin, 1996, Goodwin et al., 1992). Several methods have been developed using this approach for autoregressive models, see Fan, Söderström, Mossberg et al. (1999), Fan, Söderström, and Zhou (1999), Larsson and Söderström (2002) and Söderström and Mossberg (2000). This approach is advantageous in many cases as it is computationally efficient and avoids nonlinear transformations if the underlying model is autoregressive. This technique also benefits from nonuniform sampling (Larsson, 2004, Larsson and Söderström, 2002). However, it is not well understood how we can extend this technique for an ARMA model since the mapping from the continuous-time zeros to the equivalent discrete-time zeros is complicated. It is also not known how to guarantee the positivity of the estimated spectrum. Therefore, for ARMA models, theonly available time domain approach seems to be the indirect method described above (Larsson and Mossberg, 2003, Söderström, 1991).
If we consider a discrete-time process, we can ensure the stability of the estimated autoregressive model in a fairly simple way (Ljung, 1999, Makhoul, 1975, Söderström and Stoica, 1989). But unlike the discrete-time counterpart, the mapping from the lagged covariance estimates to the system parameters for a continuous-time system is more complicated. Hence, the standard discrete-time algorithms cannot be extended directly. In this paper, we propose a direct approach for modeling continuous-time stochastic processes. Specifically, we provide an estimation algorithm with the following properties: (i) it is computationally efficient; (ii) the stability of the estimated model is guaranteed; (iii) it can handle irregularly sampled data; (iv) it is possible to circumvent the problems associated with the sampling zeros for an ARMA model.
Our approach uses the framework of input-to-state filtering (Georgiou, 2001, Georgiou, 2002a, Georgiou, 2002b) where we first estimate the half-spectrum and its derivatives evaluated at some pre-specified points in the right-half plane. This is achieved using a linear operation on the covariance matrix of the output of an input-to-state filter. Subsequently, we present an approach for estimating a stable rational model from the estimates of the half-spectrum. In this step, we apply linear interpolation combined with a regularization step similar to that proposed in Mari, Stoica, and McKelvey (2000) and Stoica, McKelvey, and Mari (2000). If the resulting model is unstable then a recent spectral zero assignment algorithm (Byrnes, Georgiou, & Lindquist, 2001), (see also Georgiou, 1999), is used to compute a stable model. The approach can also be used when the data are irregularly sampled. We provide additional insights into the estimation of the half-spectrum, and discuss the numerical and statistical issues involved in the estimation of such statistics. We also carry out an asymptotic second-order statistical analysis. The proposed algorithm is tested using numerical simulations.
In the framework of input-to-state filtering it is also possible to model the spectrum in terms of the generalized orthogonal basis functions (Heuberger, Van den Hof, & Wahlberg, 2005). This approach has been explored in Blomqvist and Fanizza (2003).
Section snippets
Input-to-state filters
In this section, we briefly state a few key results for any continuous-time wide-sense stationary stochastic process . As a special application, we will apply these results to continuous-time ARMA processes in the later sections. Analogous results for discrete-time processes have been derived in Georgiou (2001, Theorem 1), Georgiou (2002a) and Georgiou (2002b, Corollary 1). The extensions for continuous-time case can be found in Georgiou (2002a, Section V).
Consider a scalar and real-valued
Some computational and statistical aspects
In this section, we focus on the computational and the statistical aspects involved in the estimation of the half-spectrum. The primary aim is to justify the algorithm for computing and proposed in Section 2. First, we examine the rank of the system of equations to be solved in order to determine . We also comment on the choice of the coordinates of the state of the input-to-state filter. Finally, we determine the second-order statistics of and . To this end, we need some
Fitting a rational model
The algorithm at the end of Section 2 gives the estimates of at the interpolation points. In this section, we propose an approach to fit a rational model to the interpolation data. Assume that has a strictly proper rational spectrum of order , i.e.,where Then the half-spectrum also admits a strictly proper rational representation as such thatOur approach
An illustrative example
In this section, we illustrate the proposed direct modeling approach using numerical simulations. To conduct the simulation, we first need to simulate the sampled version of a continuous-time stochastic process . We do so by using the method in Larsson (2004) and Söderström (2002) as follows. First, we express the continuous-time process in state space:where is the state vector, and is a unity variance continuous-time white noise, i.e.,
Conclusions
In this paper, we have proposed a novel direct approach for modeling continuous-time stochastic processes. The main idea is to use an input-to-state filter to compute the half-spectrum in some prescribed points in the right-half plane. The estimated samples of the half-spectrum are then used to obtain a rational model of the half-spectrum using linear interpolation with a positivity constraint. This is done by solving a semidefinite programming problem. The unique feature of the approach is
Kaushik Mahata received his M.E. degree in Signal Processing from Indian Institute of Science, Bangalore, India in 2000, and a Ph.D. degree in Signal Processing from Uppsala University, Sweden in 2003. Currently, he is a research academic in the University of Newcastle, Australia. His research interest includes estimation identification, and spectrum analysis.
References (31)
- et al.
Identification of continuous-time AR processes from unevenly sampled data
Automatica
(2002) - et al.
Performance evaluation of methods for identifying continuous-time autoregressive processes
Automatica
(2000) The effect of rapid sampling in system identification
Automatica
(1990)- Blomqvist, A., & Fanizza, V. (2003). Identification of rational spectral densities using orthonormal basis functions....
- et al.
A generalized entropy criterion for Nevanlinna–Pick interpolation with degree constraint
IEEE Transactions on Automatic Control
(2001) - et al.
Speech modeling and the trigonometric moment problem
Philips Journal of Research
(1982) - et al.
Estimation of continuous-time AR process parameters from discrete-time data
IEEE Transactions on Signal Processing
(1999) - Fan, H., Söderström, T., & Zhou, Y. (1999). Continuous-time AR process parameter estimation in presence of additive...
- et al.
Sampling in digital signal processing and control
(1996) Design with operational amplifiers and analog integrated circuits
(2001)
The interpolation problem with a degree constraint
IEEE Transactions on Automatic Control
Spectral estimation via selective harmonic amplification
IEEE Transactions on Automatic Control
Spectral analysis based on the state covariance: The maximum entropy spectrum and linear fractional parametrization
IEEE Transactions on Automatic Control
The structure of state covariances and its relation to the power spectrum of the input
IEEE Transactions on Automatic Control
Cited by (7)
On the indirect approaches for CARMA model identification
2007, AutomaticaCitation Excerpt :However, the extension of this technique to continuous-time ARMA (CARMA) models is nontrivial. An alternative analytic interpolation framework is proposed in Mahata and Fu (2006), but it is not clear how to extend this algorithm for multivariable processes. In this paper, the indirect approach is examined in detail.
Likelihood Analysis of Power Spectra and Generalized Moment Problems
2017, IEEE Transactions on Automatic ControlSome computational aspects of Gaussian CARMA modelling
2015, Statistics and ComputingNonlinear control for magnetic levitation systems based on fast online algebraic identification of the input gain
2011, IEEE Transactions on Control Systems TechnologyAdaptive controller for single-link flexible manipulators based on algebraic identification and generalized proportional integral control
2009, IEEE Transactions on Systems, Man, and Cybernetics, Part B: CyberneticsLinear LMS compensation for timing mismatch in time-interleaved ADCs
2009, IEEE Transactions on Circuits and Systems I: Regular Papers
Kaushik Mahata received his M.E. degree in Signal Processing from Indian Institute of Science, Bangalore, India in 2000, and a Ph.D. degree in Signal Processing from Uppsala University, Sweden in 2003. Currently, he is a research academic in the University of Newcastle, Australia. His research interest includes estimation identification, and spectrum analysis.
Minyue Fu received his Bachelor's Degree in Electrical Engineering from the University of Science and Technology of China, Hefei, China, in 1982, and M.S. and Ph.D. degrees in Electrical Engineering from the University of Wisconsin-Madison in 1983 and 1987, respectively.
From 1987 to 1989, he served as an Assistant Professor in the Department of Electrical and Computer Engineering, Wayne State University, Detroit, Michigan. For the summer of 1989, he was employed by the Universite Catholoque de Louvain, Belgium, as a Maitre de Conferences Invited. He joined the Department of Electrical and Computer Engineering, the University of Newcastle, Australia, in 1989. Currently, he is a Chair Professor in Electrical Engineering and Head of School of Electrical Engineering and Computer Science. He was a Visiting Associate Professor at University of Iowa in 1995–1996 and a Visiting Professor at Nanyang Technological University, Singapore, 2002.
His main research interests include control systems, signal processing and communications. He has been an Associate Editor for the IEEE Transactions on Automatic Control, Automatica, and Journal of Optimization and Engineering. He is an Fellow of IEEE.
- ☆
This paper was presented at the IFAC World Congress 2005. This paper was recommended for publication in revised form by Associate Editor Michel Verhaegen under the direction of Editor Torsten Soederstroem. The research is supported by Australian research council.