Square-root array forms of the minimum output energy detector and estimator

Computation of the minimum output energy (MOE) detector and estimator is considered. Different ways to compute the detector and estimator are reviewed. Emphasis is on square-root factorization based methods, which compute the detector and estimator iteratively, and are numerically robust. In particular, explicit algorithms based on the Householder transformation are derived. As an intermediate step, the iterative square-root algorithm to update the inverse of the sample correlation matrix is derived. This algorithm has a complexity N² operations per N×1 sample vector, and it avoids divisions and square-roots almost completely