Design Of Input Sequence Research Articles

Informative input design for Kernel-Based system identification

This paper discusses the design of input sequence for Kernel-Based system identification. From the Bayesian point of view, the kernel reflects a priori information about the target system, which implies that the information obtained from I/O data differs over kernels. This paper focuses on finding an input sequence which maximizes the information obtained through an observation according to the kernel which is given in advance. As an appropriate measure of such information, the mutual information is adopted. For the given kernel, a concrete procedure is proposed to find the input sequence maximizing the mutual information subject to the input energy constraints. Numerical examples are given to illustrate the effectiveness of the proposed input design. Furthermore, it is shown analytically that the impulse input is optimal for a special class of kernels.

Optimal input design for set-membership identification of Hammerstein models

In this paper the problem of optimal input design for the identification of Hammerstein models is considered under the assumption that the linear dynamic part of the model is a FIR and that lower and upper bounds are available for the additive measurement errors. The parameters of the Hammerstein model can then be estimated via the identification of a linearized augmented Hammerstein model . External approximations of the feasible intervals for the parameters of the original Hammerstein models are then derived (which may correspond to the actual feasible intervals). This paper deals with the design of input sequences minimizing parameter uncertainty for the linearized augmented Hammerstein model . Some new results are also reported about optimal input design for polynomial non-linear blocks, that may be part of Hammerstein models.

Fault-detection experiment with deterministic realizations and a nondeterministic model

The algorithmic approach to the design of fault-detection experiments for testing of control devices requires a mathematical model that describes the behavior of the tested device and its reference model. A finite automaton usually provides a suitable mathematical model. The classical problem assumes that the reference model and the tested device behave as a deterministic f'mite automaton (in what follows, we use the term "automaton" for a deterministic f'mite automaton). The class of faults is limited to faults that do not increase the number of automaton states. It is also assumed that the experimenter has direct access to the inputs and outputs of the device being tested. The construction of multiple fault-detection experiments assumes the existence of an input symbol that allows the model automaton to return from any state to some fixed (initial) state. This special symbol remains a reset symbol under all faults [1]. Recent studies use a nondeterministic automaton (nd-automaton) as a reference model for testing. In particular, an ndautomaton is used to construct test cases that check computer network protocols for conformity [2-4]. It has been shown [5] that if the output of the device being tested is observable only on the output of another device, then a fault,detection experiment also can be constructed using a nondeterministic model. The construction of a fault-detection experiment usually focuses mainly on the design of input sequences. The set of admissible output responses is identified with the entire set of model output sequences. In this article we analyze the possibility of reducing the length of the fault-detection experiment by reducing the set of admissible output responses. The article is organized as follows. Section 1 introduces the main def'mitions and notation. Section 2 describes the possibility of reducing the admissible set of output sequences in a fault-detection experiment. Section 3 provides necessary and sufficient conditions for the realization of a nondeterministic automaton by a deterministic automaton. We examine the conditions that permit passing from the nd-automaton P to an nd-automaton P,~ with fewer states and fewer output responses. The set of automata realizing P is identical with the set of automata realizing P~.. We can thus construct a fault~etection experiment for the nd-automaton P based on the nd-automaton P~.. This possibility is illustrated in Section 4.

Design of input sequence for linear dynamic system identification

The design of stochastic input sequences for parameter identification of linear dynamic discrete-time system is considered. The determinant of the normalized error covariance matrix is used as a criterion for parameter identification quality which is, as it is shown, dependent on the input autocorrelation function. For a given form of this function, one may derive the optimal values of its parameters. An analytic example illustrating the method is given.