Asymptotically Efficient Recursive Identification of FIR Systems With Binary-Valued Observations

doi:10.1109/TSMC.2019.2916022

	Asymptotically Efficient Recursive Identification of FIR Systems With Binary-Valued Observations
	Zhang, Hang 1,2; Wang, Ting 2; Zhao, Yanlong 1,2
刊名	IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
	2021-05-01
卷号	51 期号:5 页码:2687-2700
关键词	Finite impulse response filters Convergence Approximation algorithms Adaptive control Projection algorithms Estimation Control systems Asymptotic efficiency binary-valued observations convergence Cramé r– Rao (CR) lower bound identification
ISSN号	2168-2216
DOI	10.1109/TSMC.2019.2916022
英文摘要	This paper considers the identification problem of finite impulse response (FIR) systems with binary-valued observations under the assumption of fixed threshold and bounded persistently excitations. A recursive projection algorithm is constructed to estimate the unknown parameter. For first-order FIR systems, the convergence properties of the algorithm are analyzed theoretically. With mild conditions on the weight coefficients in the parameter update, the algorithm is proved to be convergent in mean square and the convergence rate can be the reciprocal of the number of observations, which has the same order as the optimal estimation when the system output is exactly known. Furthermore, it is also shown that the Cramer-Rao (CR) lower bound is achieved asymptotically with proper weight coefficients, which indicates that the algorithm is optimal in the sense of asymptotic efficiency. Some numerical examples are simulated to demonstrate the effectiveness of the proposed algorithm in both first-order and high-order FIR systems.
资助项目	National Key Research and Development Program of China[2016YFB0901902] ; National Natural Science Foundation of China[61622309] ; National Natural Science Foundation of China[61803370]
WOS研究方向	Automation & Control Systems ; Computer Science
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:000640749000005
内容类型	期刊论文
源URL	[http://ir.amss.ac.cn/handle/2S8OKBNM/58481]
专题	中国科学院数学与系统科学研究院
通讯作者	Zhao, Yanlong
作者单位	1.Univ Chinese Acad Sci, Sch Math Sci, Beijing 100049, Peoples R China 2.Chinese Acad Sci, Acad Math & Syst Sci, Key Lab Syst & Control, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Zhang, Hang,Wang, Ting,Zhao, Yanlong. Asymptotically Efficient Recursive Identification of FIR Systems With Binary-Valued Observations[J]. IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS,2021,51(5):2687-2700.
APA	Zhang, Hang,Wang, Ting,&Zhao, Yanlong.(2021).Asymptotically Efficient Recursive Identification of FIR Systems With Binary-Valued Observations.IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS,51(5),2687-2700.
MLA	Zhang, Hang,et al."Asymptotically Efficient Recursive Identification of FIR Systems With Binary-Valued Observations".IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS 51.5(2021):2687-2700.