Hybrid Beamforming in Massive-MIMO mmWave Systems Using LU Decomposition

Beamforming is considered as one of the most important enabling techniques for millimeter wave (mmWave) with massive multiple-in-multiple-out (MIMO) communications systems. Fully digital beamforming has been proven to be infeasible from the hardware point of view. Analog Beamforming, on the other hand, can not achieve as good spectral efficiency as required for the high throughput applications expected for the next generation of wireless communications. So, the hybrid beamforming, as a combination of digital baseband and analog RF phase shifters, has been considered the common sense when designing beamforming systems. Achieving a performance that is close to the fully digital beamformers has been the goal of many studies recently. In this paper, we suggest a simple, yet effective way of designing the digital and analog parts of the beamformers by decomposing the Fully digital optimal beamforming matrix into two matrices for the digital and the analog parts of the hybrid beamformer using the matrix factorization method called LU-Decomposition. Then we test the results to make sure they obey the hardware limitations of the system. Our results shows that it is possible to build a very efficient hybrid beamformer with such simple method and to get as close performance to the fully digital one as possible.