Class Of System Identification Problems Research Articles

Neural Membrane Mutual Coupling Characterisation Using Entropy-Based Iterative Learning Identification

This paper investigates the interaction phenomena of the coupled axons while the mutual coupling factor is presented as a pairwise description. Based on the Hodgkin-Huxley model and the coupling factor matrix, the membrane potentials of the coupled myelinated/unmyelinated axons are quantified which implies that the neural coupling can be characterised by the presented coupling factor. Meanwhile the equivalent electric circuit is supplied to illustrate the physical meaning of this extended model. In order to estimate the coupling factor, a data-based iterative learning identification algorithm is presented where the Rényi entropy of the estimation error has been minimised. The convergence of the presented algorithm is analysed and the learning rate is designed. To verified the presented model and the algorithm, the numerical simulation results indicate the correctness and the effectiveness. Furthermore, the statistical description of the neural coupling, the approximation using ordinary differential equation, the measurement and the conduction of the nerve signals are discussed respectively as advanced topics. The novelties can be summarised as follows: 1) the Hodgkin-Huxley model has been extended considering the mutual interaction between the neural axon membranes, 2) the iterative learning approach has been developed for factor identification using entropy criterion, and 3) the theoretical framework has been established for this class of system identification problems with convergence analysis.

System Identification under Information Security

There are important information security problems for system identification, which have not been paid attention to and well studied. In this paper, we discuss the security problem of input information according to a class of system identification problems, and present security protocols in the sense of cryptography. It is ensured that the input information is not leaked during solving the identification problem, which can deal with passive attacks effectively. In addition, a quantitative relationship between the input information and the number of participants is given in this paper. The simulation results show that, in the range of allowed errors, the identification algorithm with the security protocol achieves the estimation accuracy. Moreover, the corresponding time complexity increases comparing with the original identification algorithm.

Information Security Protocol Based System Identification with Binary-Valued Observations

Traditional control does not pay much attention to information security problems in system identification enough, which are important in practical applications. This paper focuses on the security problem of input information in a class of system identification problems with noise and binary-valued observations, presents a cryptography based security protocol, and improves it in the range of allowed errors. During solving the identification problem, the improved security protocol can ensure that the input information is not leaked, and thus, can deal with passive attacks effectively. Besides, a quantitative relationship among the input information, the public key in encryption and the number of participants in the improved protocol is shown. Finally, the simulation results show that, the identification algorithm can still achieve the estimation accuracy by adding the improved security protocol. However, compared with the original identification algorithm, the time complexity of the algorithm with the improved security protocol increases.

The construction of band symmetric models for vibratory systems from modal analysis data

Abstract Some natural frequencies and their corresponding mass normalized mode shapes of physical structures may be measured by experiment. Since the discretization of the continuous model of the structure reflects only a limited degree of coupling between model elements, the associated mass and stiffness matrices are typically sparse. Thus, appropriately numbered model elements may lead to banded matrices. In this paper a method is presented which determines the bandwidths of the mass and stiffness matrices, and then recovers the model by using eigendata. This approach falls somewhere between the wider class of system identification problems, in which the only structural property assumed is symmetry (and, hence, for which no unique physical model can be constructed), and inverse problems in which the model interconnections are known. The present approach thus widens the class of models that may be reconstructed.

On system identification by nonparametric function fitting

A new nonparametric algorithm for function fitting is proposed and its application to a broad class of system identification problems is discussed.

Determination of Volterra kernels from input-output data

In this paper a class of system identification problems of non-parametric type is considered. Specifically, given an unknown deterministic system whose input-output relation can be expressed by a Volterra expansion, some basic structural aspects related to the problem of determining Volterra kernels from input-output records are discussed from a coordinate-free point of view. The model adopted for the system input—output relation is general enough to cover a number of situations ranging from problems of identification where the kernels are time-varying to those where the kernels must be constrained in Hilbert spaces of well-behaved functions. The emphasis is on establishing what, in principle, can be recovered of the system kernels through a noiseless identification process, and the ultimate limitations imposed by the presence of observation or measurement noise. An application of the steepest descent iteration to the ill-posed estimation problem resulting from the presence of observation noise is considered.

A note on stochastic approximation algorithms in system identification

This correspondence considers the application of stochastic approximation algorithms to a broad class of system identification problems. Both asymptotic and initial convergence properties of the algorithms are discussed. A suboptimal procedure for parameter selection and a means of convergence acceleration are suggested.

Deterministic approach to a class of nonparametric system identification problems

In this paper a class of system identification problems of nonparametric type is considered. Specifically, given an unknown deterministic system, some basic structural aspects related to the problem of estimating its weighting function are discussed. The model adopted for the system input-output relation is general enough to cover a number of situations ranging from problems of identification of linear time-invariant systems to those where the system is nonlinear and time varying. The emphasis is on establishing what, in principle, can be recovered of the system weighting function through a noiseless identification process and the ultimate limitations imposed by the presence of observation or measurement noise.

On the structure of a class of system identification problems

Systems of the form y= Hx are considered, where x is the input signal, y is the output and H is a transformation. The input and output signals belong to L 2 spaces, H belongs to a specified class of continuous, not necessarily linear, transformations. The fundamental problem is to identify H from measurements of x and y. The approximate representation of certain transformations H in terms of polynomial integral operators is discussed. The concept of a determinable class, which is used throughout, and some of its properties are reviewed. An illustrative example concerning human operator tracking problems is presented. A structural characterization of determinable classes based on work of Hsieh and Balakrishnan is given and then used as foundation for an application of linear regression analysis to the problem of identifying an H belonging to a determinable class when there is observation noise. Finally, the notion of determinable class is applied to a class of stochastic operators and an elementary convergence theorem is given.