Signal Processing in the Linear Statistical Model
Abstract
This research program has produced the following results: (1) a method of estimating power spectra using multiple orthogonal time-division windows; (2) an exact likelihood procedure for estimating frequencies and arrival angles for propagating plane waves that are detected by multisensor arrays; (3) a theory of oblique projections for solving estimation problems; (4) a method for designing matched subspace detectors for detecting subspace signals in subspace interference; (5) a method for designing data-adaptive and reduced- rank estimators for solving least squares problems; and (6) a method for bounding the probability of a subspace swap in SVD methods. Work is continuing on (1) general quadratic performance bounds for estimation; (2) complex exponential frames for modeling discrete-time sequences; (3) discrete wavelet frames and fitter banks; (4) projection operator decomposition of L2; and (5) representation and synthesis of periodically correlated time series using subband decompositions.
Document Details
- Document Type
- Technical Report
- Publication Date
- Oct 20, 1993
- Accession Number
- ADA272114
Entities
People
- Clifford T. Mullis
- Louis L. Scharf
Organizations
- University of Colorado Boulder