Topic
System identification
About: System identification is a research topic. Over the lifetime, 21291 publications have been published within this topic receiving 439142 citations.
Papers published on a yearly basis
Papers
More filters
TL;DR: In this article, an efficient open-loop digital predistorter (DPD) derived from the dynamic deviation reduction-based Volterra series that allows compensation for both nonlinear distortion and memory effects induced by RF power amplifiers in wireless transmitters is proposed.
Abstract: In this paper, we propose an efficient open-loop digital predistorter (DPD) derived from the dynamic deviation reduction-based Volterra series that allows compensation for both nonlinear distortion and memory effects induced by RF power amplifiers in wireless transmitters. In this approach, the parameters of the predistorter can be directly extracted from an offline system identification process. This eliminates the usual requirement for a closed-loop real-time parameter adaptation, which dramatically reduces the implementation complexity of the system. It is shown that a further reduction in system complexity can be achieved by applying under-sampling theory in the model extraction and utilizing parameter interpolation in the DPD implementation. Experimental results show that by utilizing this technique with only a small number of parameters, nonlinear distortion induced by the PA can be significantly reduced, as evaluated by both adjacent channel power ratio reduction and normalized root mean square error improvement. A comparison with a memoryless polynomial function based predistorter and an analysis of the impact of decresting are also presented.
266 citations
TL;DR: The results of this paper provide a unifying framework under which all these algorithms can be viewed and the link with VARX modeling have important implications as to computational complexity is concerned, leading to very computationally attractive implementations.
265 citations
TL;DR: The analysis indicates that the partially C-SG algorithm can give more accurate parameter estimates than the standard stochastic gradient (SG) algorithm.
Abstract: This technical note addresses identification problems of non-uniformly sampled systems. For the input-output representation of non-uniform discrete-time systems, a partially coupled stochastic gradient (C-SG) algorithm is proposed to estimate the model parameters with high computational efficiency compared with the standard stochastic gradient (SG) algorithm. The analysis indicates that the partially C-SG algorithm can give more accurate parameter estimates than the SG algorithm. The parameter estimates obtained using the partially C-SG algorithm converge to their true values as the data length approaches infinity.
263 citations
Proceedings Article•
01 Dec 1998TL;DR: A generalization of the EM algorithm for parameter estimation in nonlinear dynamical systems if Gaussian radial basis function (RBF) approximators are used to model the nonlinearities, the integrals become tractable and the maximization step can be solved via systems of linear equations.
Abstract: The Expectation-Maximization (EM) algorithm is an iterative procedure for maximum likelihood parameter estimation from data sets with missing or hidden variables [2]. It has been applied to system identification in linear stochastic state-space models, where the state variables are hidden from the observer and both the state and the parameters of the model have to be estimated simultaneously [9]. We present a generalization of the EM algorithm for parameter estimation in nonlinear dynamical systems. The "expectation" step makes use of Extended Kalman Smoothing to estimate the state, while the "maximization" step re-estimates the parameters using these uncertain state estimates. In general, the nonlinear maximization step is difficult because it requires integrating out the uncertainty in the states. However, if Gaussian radial basis function (RBF) approximators are used to model the nonlinearities, the integrals become tractable and the maximization step can be solved via systems of linear equations.
261 citations
Book•
20 Jun 1994TL;DR: In this article, the authors propose a deterministic LQ regulation based on Riccati-based solution via Polynomial Equations (PE) to solve the problem of linear systems.
Abstract: 1. Introduction. I. BASIC DETERMINISTIC THEORY OF LQ AND PREDICTIVE CONTROL. 2. Deterministic LQ Regulation - I: Riccati-Based Solution. 3. I/O Descriptions and Feedback Systems. 4. Deterministic LQ Regulation - II: Solution via Polynomial Equations. 5. Deterministic Receding Horizon Control. II. STATE ESTIMATION, SYSTEM IDENTIFICATION, LQ AND PREDICTIVE STOCHASTIC CONTROL. 6. Recursive State Filtering and System Identification. 7. LQ and Predictive Stochastic Control. III. ADAPTIVE CONTROL. 8. Single-Step-Ahead Self-Tuning Control. 9. Adaptive Predictive Control. APPENDICES. A. Some Results from Linear Systems Theory. B. Some Results of Polynomial Matrix Theory. C. Some Results on Linear Diophantine Equations. D. Probability Theory and Stochastic Processes. References. Some Often Used Abbreviations. Index.
261 citations