Efficient Algorithms for Computing the Capon and APES Filters

The computation of the Capon and the amplitude and phase estimation (APES) filters can be very time consuming. The burden lies not only in the inversion of the sample covariance matrix, but also in the matrix multiplications. In this paper, we propose an efficient algorithm for the computation of the above mentioned filters, using Cholesky factorization and fast Fourier transforms. Furthermore, we discuss and illustrate the benefits of using the proposed algorithms in the spectral analysis of data with missing samples.