Title :
On the numerical conditioning in orthogonal filters
Author :
Maskarinec, Gregory J. ; Chitrapu, Prabhakar R.
Author_Institution :
Dept. of Electr. & Comput. Eng., Drexel Univ., Philadelphia, PA, USA
Abstract :
Orthogonal embedding filters are defined in terms of the numerical conditioning of the realization matrix which describes the state-update and output computations in a digital filter. The construction of an orthogonal embedding filter consists of embedding a desired transfer function into a multivariable all-pass system which is implemented by an orthogonal realization matrix. The embedding system is ideally conditioned, and the numerical conditioning of the partial transfer realization matrix of interest is addressed. The numerical conditioning of the partial transfer realizations in orthogonal embedding filters is examined. Additionally, the partial transfer realizations are related to the class of generalized orthogonal state variable filters, or principle axis realizations which are defined in terms of the statistical properties of the state variables
Keywords :
all-pass filters; digital filters; matrix algebra; transfer functions; embedding filters; multivariable all-pass system; numerical conditioning; orthogonal filters; output computations; partial transfer realization matrix; principle axis realizations; realization matrix; state-update; statistical properties; transfer function; Controllability; Digital filters; Eigenvalues and eigenfunctions; Embedded computing; MIMO; Observability; Transfer functions; White noise;
Conference_Titel :
Circuits and Systems, 1993., ISCAS '93, 1993 IEEE International Symposium on
Conference_Location :
Chicago, IL
Print_ISBN :
0-7803-1281-3
DOI :
10.1109/ISCAS.1993.394228