Nonnegative Joint Diagonalization by Congruence Based on LU Matrix Factorization

摘要: In this letter, a new algorithm for joint diagonalization of set matrices by congruence is proposed to compute the nonnegative diagonalizer. The nonnegativity constraint imposed means square change variables. Then we formulate high-dimensional optimization problem into several sequential polynomial subproblems using LU matrix factorization. Numerical experiments on simulated emphasize advantages method, especially in case degeneracies such as low SNR values and small number matrices. An illustration blind separation nuclear magnetic resonance spectroscopy confirms validity improvement method.