DOA estimation based on time-frequency music application to Massive MIMO systems

In this work, the problem of direction finding is addressed. We show the Direction Of Arrival (DOA) estimation can be realized through the non-unitary joint diagonalization of spatial quadratic time-frequency. We use an approach of selection of time-frequency points to construct the set of matrices which will be jointly diagonalized to estimate the noise subspace. The main advantage of this method is that it does not require any whitening stage, and thus, it is intended to work even with a class of correlated signals. Finally, the noise subspace obtained is then used to estimate the directions using the MUltiple SIgnal Classification MUSIC spectrum. Numerical simulations are provided in order to illustrate the effectiveness and the behavior of the proposed approach.