Spatiotemporal-MIMO Channel Estimator and Beamformer for 5G

With requirements of spiraling data rates and limited spectrum availability, there is an increased interest in mm-wave beamformer-based communications for 5G. For upcoming cellular networks, the critical point is to exploit the increased number of employable antennas at both Tx and Rx to: 1) combat increased path loss; 2) tackle higher interference due to higher user density; and 3) handle multipath effects in frequency selective channels. Toward this, a multi-beam spatiotemporal superresolution beamforming framework is proposed in this paper as a promising candidate to design beampatterns that mitigate/suppress co-channel interference and deliver massive gain in the desired directions. Initially, channel and signal models suitable for the mm-wave MIMO system are presented using the manifold vectors of both Tx and Rx antenna arrays. Based on these models, a novel subspace-based channel estimator is employed, which estimates delays, directions, velocities, and fading coefficients of the desired signal paths. This information is then exploited by the proposed spatiotemporal beamformer to provide a massive array gain that combats path loss without increasing the number of antenna array elements and to be tolerant to the near-far problem in a high interference environment. The performance of the proposed channel estimator and beamformer is examined using computer simulation studies.