Title :
LTI systems, additive noise, and order estimation
Author :
Beheshti, Soosan ; Dahleh, Munther A.
Author_Institution :
Dept. of Electr. Eng. & Comput. Sci., Massachusetts Inst. of Technol., Cambridge, MA, USA
Abstract :
This paper presents an important application of a novel information theoretic order estimation method, minimum description complexity (MDC). The selection of optimum number of poles and zeros in identification of LTI systems based on observed data is accomplished by MDC. The comparison of MDC with important existing order estimation methods, MDL and AIC, is provided.
Keywords :
computational complexity; estimation theory; identification; noise; poles and zeros; LTI systems; additive noise; linear time invariant system; minimum description complexity; order estimation; poles and zeros; system identification; AWGN; Additive noise; Additive white noise; Application software; Estimation theory; Laboratories; Maximum likelihood estimation; Parametric statistics; Poles and zeros; Random processes;
Conference_Titel :
Decision and Control, 2003. Proceedings. 42nd IEEE Conference on
Print_ISBN :
0-7803-7924-1
DOI :
10.1109/CDC.2003.1272392
