Novel transceiver beamforming schemes for a MIMO-GFDM system

Generalized frequency division multiplexing (GFDM) is a multicarrier system to deal with highly dispersive channels by well-localized non-orthogonal pulse shapes. In order to achieve spatial diversity and enhance the performance, multiple input multiple output (MIMO) beamforming can be adapted to GFDM. Since the transmitted symbols are overlapped in time and frequency, the conventional beamforming techniques cannot be used straightforwardly in MIMO-GFDM. In this paper, three novel schemes are proposed for beamforming of the transceiver antennas in a MIMO-GFDM system, based on the singular value decomposition (SVD) technique. The total power of the noise after symbol estimation is considered as a criterion to design the beamforming matrices. The computational complexities of the proposed MIMO-GFDM beamforming schemes are evaluated and the corresponding bit error rate (BER) performances are compared with a well-known SVD based beamforming in MIMO orthogonal frequency division multiplexing (MIMO-OFDM) via simulation. We show that as a trade off between the computational complexity and the BER performance, one of these proposed schemes can be chosen for data transmission. (C) 2021 Elsevier B.V. All rights reserved.

Automated summary

GFDM is a multicarrier system designed to handle highly dispersive channels using well-localized non-orthogonal pulse shapes, with MIMO beamforming techniques to enhance spatial diversity and performance; three novel schemes for beamforming in a MIMO-GFDM system based on SVD technique are proposed, considering noise power as a criterion for designing beamforming matrices; computational complexities and BER performances are evaluated to find a balance between computational complexity and performance.