Extraction of an eigenvector-oriented "Spectrum" from the MESA coefficients

In this communication, an efficient algorithm is proposed for extracting an orthogonal eigenvector-oriented spectrum directly from maximum entropy spectral analysis (MESA) coefficients. In effect, the algorithm estimates the eigenvector corresponding to the smallest eigenvalue of the data covariance matrix. Simulations show that in a narrow-band array processing application, the multisource resolution of MESA can be significantly improved with the use of this algorithm.