scholarly journals An Algorithm to Generate Representations of System Identification Errors

2008 ◽  
Vol 2008 ◽  
pp. 1-4 ◽  
Author(s):  
Wancheng Zhang ◽  
Patrick A. Naylor
Keyword(s):  
System Identification ◽  
Impulse Response ◽  
Scaling Factors ◽  
Systematic Testing ◽  
Normalized Projection

An algorithm to generate representations of system identification (SI) errors, which enables systematic testing of the performance of system equalization techniques, is proposed. With this algorithm, the normalized projection misalignment (NPM) of the generated error representation can be chosen to suit the particular characteristics of the application under test. Additionally, the generated error representation can represent all the error vectors corresponding to different scaling factors in the estimates of the system impulse response (SIR), without influencing the signal-to-distortion ratio (SDR) of the equalized impulse response.

An Improved ZA-LMS Algorithm for Sparse System Identification with Variable Sparsity

2014 ◽  
Vol 602-605 ◽  
pp. 2415-2419 ◽  
Author(s):  
Hui Luo ◽  
Yun Lin ◽  
Qing Xia
Keyword(s):  
System Identification ◽  
Impulse Response ◽  
Lms Algorithm ◽  
Mean Square ◽  
Least Mean Square ◽  
Sparse System ◽  
Sparse System Identification ◽  
Least Mean Square Algorithm ◽  
Improved Algorithm

The standard least mean square algorithm does not consider the sparsity of the impulse response,and the performs of the ZA-LMS algorithm deteriorates ,as the degree of system sparsity reduces or non-sparse . Concerning this issue ,the ZA-LMS algorithm is studied and modified in this paper to improve the performance of sparse system identification .The improved algorithm by modify the zero attraction term, which attracts the coefficients only in a certain range (the “inactive” taps), thus have a good performance when the sparsity decreases. The simulations demonstrate that the proposed algorithm significantly outperforms then the ZA-LMS with variable sparisity.

Identification of Structural Dynamics Models Using Wavelet-Generated Impulse Response Data

1995 ◽  
Author(s):  
A. N. Robertson ◽  
K. C. Park ◽  
K. F. Alvin
Keyword(s):  
System Identification ◽  
Impulse Response ◽  
Structural Dynamics ◽  
Wavelet Transforms ◽  
Singular Values ◽  
Mode Shapes ◽  
Discrete Wavelet ◽  
Impulse Response Functions ◽  
Structural System Identification ◽  
Dynamics Models

Abstract This paper addresses the use of discrete wavelet tranforms for the identification of structural dynamics models. First, the discrete temporal impulse response functions are obtained from vibration records by the discrete wavelet transforms (DWTs), which are then utilized for system realizations. From the realized state space models, structural modes, mode shapes and damping parameters are extracted. Attention has been focused on a careful comparison of the present DWT system identification approach to the FFT-based approach and a rational criterion for truncating realized singular values. Numerical examples demonstrate that the present DWT-based structural system identification procedure outperforms the FFT-based procedure.

Sign in / Sign up

Export Citation Format

Share Document