Deconvolution/identification techniques for nonnegative signals

Conference · Fri Nov 01 00:00:00 EST 1991

OSTI ID:10112087

Goodman, D M; Yu, D R

Several methods for solving the nonparametric deconvolution/identification problem when the unknown is nonnegative are presented. First we consider the constrained least squares method and discuss three ways to estimate the regularization parameter: the discrepancy principle, Mallow`s C{sub L}, and generalized cross validation. Next we consider maximum entropy methods. Last, we present a new conjugate gradient algorithm. A preliminary comparison is presented; detailed Monte-Carlo experiments will be presented at the conference. 13 refs.

• Research Organization: Lawrence Livermore National Lab., CA (United States)
• Sponsoring Organization: USDOE, Washington, DC (United States)
• DOE Contract Number: W-7405-ENG-48
• OSTI ID: 10112087
• Report Number(s): UCRL-JC-108920; CONF-920354-3; ON: DE92005244
• Resource Relation: Conference: 1992 Institute of Electrical and Electronic Engineers (IEEE) international conference on acoustics, speech and signal processing,San Francisco, CA (United States),23-26 Mar 1992; Other Information: PBD: Nov 1991
• Country of Publication: United States
• Language: English

Related Subjects

99 GENERAL AND MISCELLANEOUS//MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE
47 OTHER INSTRUMENTATION
SIGNAL CONDITIONING
ANALYTICAL SOLUTION
LEAST SQUARE FIT
ALGORITHMS
MATHEMATICAL OPERATORS
MATRICES
MATHEMATICAL MODELS
NOISE
990200
440000
MATHEMATICS AND COMPUTERS
INSTRUMENTATION