Work is currently underway to design beam allocation methods for MIMO systems based on ML that outperform common conventional methods. In addition, beamforming techniques have been proposed that can optimize the number of active radio chains depending on the current channel conditions in order to maximize energy or spectral efficiency. For example, principal component analysis (PCA) and singular value decomposition (SVD) methods are used to manipulate the receive matrix of massive MIMO to reduce the computational complexity of MIMO[O42].