Exact pdfs for sample covariance based array processors with elliptically contoured data

Christ D. Richmond

Research output: Contribution to journalConference articlepeer-review


Practical application of array processors typically requires use of a sample covariance matrix (SCM). We add to the many results on SCM based (SCB) array processors by weakening the traditional assumption of data Gaussianity and subsequently providing for a class of array processors additional performance measures of value in practice. The snapshot data matrix is assumed complex multivariate elliptically contoured (MEG) distributed. The performance measures include the exact probability density functions (pdfs) and moments of the SCB weightings and beam responses of the following array processors: (1) Maximum-Likelihood signal vector estimator, (2) Linearly Constrained Minimum Variance beamformer (LCMV), (3) Minimum Variance Distortionless Response beamformer, and (4) Generalized Sidelobe Cancellor implementation of the LCMV beamformer. The SCB weightings for these array processors are complex multivariate standardized t-distributed and the SCB beam responses are generalized t. All results are completely invariant over the class of MEC's considered.

Original languageEnglish (US)
Pages (from-to)2849-2851
Number of pages3
JournalICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
StatePublished - Jan 1 1996
Externally publishedYes
EventProceedings of the 1996 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP. Part 1 (of 6) - Atlanta, GA, USA
Duration: May 7 1996May 10 1996

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Electrical and Electronic Engineering


Dive into the research topics of 'Exact pdfs for sample covariance based array processors with elliptically contoured data'. Together they form a unique fingerprint.

Cite this