Input signal design for system identification: a comparative analysis

Abstract In this paper the problem of input signal design for system identification is considered. The input signal is amplitude-constrained and the case of a finite-length Input sequence Is Investigated. Various optiraality criteria are used to derive the optimal identifying input sequence. In particular: Rissanen's Minimum Description length (MHL) criterion, the determinant of the information matrix, the information matrix determinant ratio, and the trace of the parameter error covariance matrix are considered as the criteria for optimality of identification. These input design criteria can also be used in structure selection (multivariate systems) or in model order (univariate systems) estimation procedures. The emphasis is pur on autoregressive models, and the input signal sequences are derived using a dynamic programming approach. A comparative study based on analytical results as well as on simulations is presented.

[1]  Michel Gevers,et al.  On the problem of structure selection for the identification of stationary stochastic processes , 1982 .

[2]  Anthony J. Jakeman,et al.  An instrumental variable method for model order identification , 1980, Autom..

[3]  G. Elkobrosy.  On Input Signal Synthesis for Parameter Estimation with Output Power Constraints , 1982 .

[4]  Torsten Bohlin,et al.  Experiment design for maximum-power model validation , 1980, Autom..

[5]  T. Söderström,et al.  A useful input parameterization for optimal experiment design , 1982 .

[6]  Design of input sequence for linear dynamic system identification , 1983 .

[7]  R. Guidorzi,et al.  The range error test in the structural identification of linear multivariable systems , 1982 .

[8]  Tung-Sang Ng,et al.  Optimal experiment design for linear systems with input-output constraints , 1977, Autom..

[9]  Lennart Ljung,et al.  Identification of processes in closed loop - identifiability and accuracy aspects , 1977, Autom..

[10]  J. Rissanen.  A UNIVERSAL PRIOR FOR INTEGERS AND ESTIMATION BY MINIMUM DESCRIPTION LENGTH , 1983 .

[11]  J. Hrušák,et al.  Accuracy Aspects of Parameter Estimation in Linear Dynamical Systems , 1982 .

[12]  Tung-Sang Ng,et al.  Optimal input design for an AR model with output constraints , 1984, Autom..

[13]  J. Rissanen.  Estimation of structure by minimum description length , 1982 .

[14]  T. Ng,et al.  Optimal experiment design for autoregressive model with output power constraints , 1981 .

[15]  S. Arimoto,et al.  Optimum input test signals for system identification—an information-theoretical approach , 1971 .

[16]  M. Ljubojević,et al.  Suboptimal input signals for linear system identification , 1973 .

[17]  Robert E. Kalaba,et al.  Optimal inputs and sensitivities for parameter estimation , 1973 .

[19]  Belle R. Upadhyaya,et al.  Synthesis of linear stochastic signals in identification problems , 1976, 1976 IEEE Conference on Decision and Control including the 15th Symposium on Adaptive Processes.

[20]  Katsuji Uosaki,et al.  Optimal Input Design for Autoregressive Model Discrimination with Output Amplitude Constraints , 1984 .

[21]  R. Mehra.  Optimal inputs for linear system identification , 1974 .

[22]  M. Athans,et al.  Optimal policies for identification of stochastic linear systems , 1975 .

[23]  Raman K. Mehra,et al.  Optimal input signals for parameter estimation in dynamic systems--Survey and new results , 1974 .

[24]  C. Berger.  Synthesis of input signals for parameter identification using moving-average filter , 1979 .

[25]  R. M. Staley,et al.  On input signal synthesis in parameter identification , 1970 .

[26]  B. Upadhyaya.  Characterization of optimal inputs by generating functions for linear system identification , 1980 .