Title :
Conjugate gradient eigenstructure tracking for adaptive spectral estimation
Author :
Fu, Zuqiang ; Dowling, Eric M.
Author_Institution :
Erik Jonsson Sch. of Eng. & Comput. Sci., Texas Univ., Richardson, TX, USA
fDate :
5/1/1995 12:00:00 AM
Abstract :
A conjugate gradient iteration is derived that converges to the set of r dominant/subdominant eigenpairs. This iteration is used to construct two eigenstructure tracking algorithms that track the r-dimensional dominant or subdominant subspaces of time-varying data or data-covariance matrices. The two eigenstructure tracking algorithms have update complexities O(m2r) and the other O(mr2), where m is the data dimension. The algorithms are customized to solve high resolution temporal and spatial frequency tracking problems. They are compared with existing techniques by tying into published simulation based performance tests. The algorithms demonstrate rapid convergence and tracking characteristics at a competitive cost
Keywords :
adaptive signal processing; conjugate gradient methods; convergence of numerical methods; covariance matrices; direction-of-arrival estimation; eigenstructure assignment; signal resolution; spectral analysis; tracking; DOA estimation; adaptive spectral estimation; conjugate gradient eigenstructure tracking; convergence; cost function; data dimension; data-covariance matrix; dominant/subdominant eigenpairs; eigenstructure tracking algorithms; eigenvalue decomposition; high resolution spatial frequency tracking; high resolution temporal frequency tracking; iteration; performance tests; simulation; time-varying data matrix; tracking characteristics; update complexities; Costs; Covariance matrix; Eigenvalues and eigenfunctions; Frequency estimation; Matrix decomposition; Multiple signal classification; Signal processing algorithms; Spatial resolution; Symmetric matrices; Testing;
Journal_Title :
Signal Processing, IEEE Transactions on
