Graduate Seminar

Graduate Seminar"Subspace Identification Using the Parity Space"Jin Wang, Ph.D., P.E.
Advanced Micro Devices, IncABSTRACT Subspace identification methods (SIMs) have been one of the main streams of research in system identification. Compared to the prediction error methods (PEMs), SIMs have a better numerical reliability and a modest computational complexity, particularly when the number of outputs and states is large. However, most of the SIMs, like other more traditional PEMs, consider output errors only and assume the input variables are noise-free. Therefore, under the errors-in-variables (EIV) situation, most of the existing SIMs gives biased estimates. Besides, due to the correlation between the input and the unmeasured disturbance under feedback control, many subspace algorithms do not work on closed-loop data, even though the data satisfy identifiability conditions for prediction error methods.
In this talk, I will present a new subspace identification method using the parity space employed in fault detection in the past. The basic algorithm, known as subspace identification method via principal component analysis (SIMPCA), gives consistent estimation of the deterministic part and stochastic part of the system, for both closed-loop and errors-in-variables situation. Two modifications, SIMPCA with column weighting and SIMPCA with modified instrumental variables, are developed to further improve the efficiency/accuracy of SIMPCA. Simulation examples are given to illustrate the performance of the proposed algorithms.Tuesday, January 24, 2006
Seminar at 12:00 p.m.
SGM 101
The Scientific Community is Cordially Invited

Location: Seeley G. Mudd Building (SGM) - 101

Audiences: Everyone Is Invited

Contact: Petra Pearce