Title :
Projections on unstructured subspaces
Author :
Krim, H. ; Forster, P.
Author_Institution :
Lab. for Inf. & Decision Syst., MIT, Cambridge, MA, USA
fDate :
10/1/1996 12:00:00 AM
Abstract :
Orthogonal projection on vector subspaces arises in many applied fields. The common assumption about the orthogonal complementary subspace is that it is spanned by white noise components. We generalize a previous perturbation analysis of projection operators to that with a noise field with an arbitrarily structured covariance matrix. The resulting expressions are insightful, and their algebraic power is very useful for applications
Keywords :
array signal processing; covariance matrices; mathematical operators; perturbation techniques; spectral analysis; white noise; algebraic power; arbitrarily structured covariance matrix; noise field; orthogonal complementary subspace; orthogonal projection; perturbation analysis; projection operators; unstructured subspaces; vector subspaces; white noise; Adaptive arrays; Electrons; Gaussian noise; Gaussian processes; H infinity control; Nonlinear filters; Signal processing; Signal to noise ratio; Statistics; White noise;
Journal_Title :
Signal Processing, IEEE Transactions on
