Spectral and sensor array analysis using maximum entropy with new eigenstructure constraints

A number of modern spectral estimators are shown to have a common generic formulation. These include minimum variance, MUSIC, maximum entropy, and their enhanced versions. A new maximum entropy spectral estimator is derived using constraints on the modal powers or the expected-square projections of the data onto the eigenvectors of the data covariance matrix. It is conjectured that these constraints may be more suitable than those of conventional maximum entropy. A formulation is given that allows incorporation of uncertainty into the modal power constraints. The resulting estimators have quite reasonable forms, and similarities to previous estimators are discussed. It is noted that the new estimators allow further development when various a priori information can be incorporated into the constraints