Title :
Comparisons of stochastic gradient and least squares algorithms for multivariable systems
Author :
Liao, Yuwu ; Liu, Yanjun ; Feng Ding
Author_Institution :
Dept. of Phys. & Electron. Inf. Technol., Xiangfan Univ., Xiangfan, China
Abstract :
Two identification models are obtained for multivariable ARX systems by different parameterization, and the corresponding two least squares and two stochastic gradient algorithms are given based on the lest squares principle and the stochastic gradient search principle and minimizing different cost functions. The performances of these algorithms are analyzed and compared by the simulation tests.
Keywords :
gradient methods; least mean squares methods; multivariable control systems; stochastic processes; least squares algorithms; multivariable ARX systems; parameter estimation; stochastic gradient algorithms; Algorithm design and analysis; Least squares methods; MIMO; Parameter estimation; Performance analysis; State estimation; State-space methods; Stochastic processes; Stochastic systems; Transfer functions; Least Squares; Multivariable Systems; Parameter Estimation; Recursive Identification; Stochastic Gradient;
Conference_Titel :
Control and Decision Conference (CCDC), 2010 Chinese
Conference_Location :
Xuzhou
Print_ISBN :
978-1-4244-5181-4
Electronic_ISBN :
978-1-4244-5182-1
DOI :
10.1109/CCDC.2010.5498589
