Coordinated beamforming in MIMO FBMC/OQAM systems

Abstract

In this contribution, we propose a coordinated transmit beamforming technique for point-to-point multiple-input-multiple-output (MIMO) filter bank based multi-carrier with offset quadrature amplitude modulation (FBMC/OQAM) systems. To enable reliable transmissions when the number of transmit antennas does not exceed the number of receive antennas and the channel is not flat fading, we design a joint and iterative procedure to calculate the precoding matrix and the decoding matrix for each subcarrier. Simulation results show that the proposed algorithm outperforms the existing transmission strategies for MIMO FBMC/OQAM systems. It is also observed that by employing the proposed coordinated beamforming scheme, the MIMO FBMC/OQAM system achieves a similar bit error rate (BER) performance as its orthogonal frequency division multiplexing with the cyclic prefix insertion (CP-OFDM) based counterpart while exhibiting superiority in terms of a higher spectral efficiency, a greater robustness against synchronization errors, and a lower outof-band radiation.Index Terms-MIMO FBMC/OQAM, coordinated beamforming, frequency selective channel

Towards a non-error floor multi-stream beamforming design for FBMC/OQAM

“…The drawback of projecting onto the null space ofĤ q is that the spatial channel gains are dramatically reduced as if no projection was needed. The ZF only gives satisfactory performance if S = N R ≤ N T , as [4], [5], [7] show. The problem exhibited by the previous design is solved to some extent by the coordinated beamforming solution [5], by setting these new constraints A T qĤqBq = 0 and forcing equalizers to be real-valued.…”

“…The ZF only gives satisfactory performance if S = N R ≤ N T , as [4], [5], [7] show. The problem exhibited by the previous design is solved to some extent by the coordinated beamforming solution [5], by setting these new constraints A T qĤqBq = 0 and forcing equalizers to be real-valued. Note that if an initial set of equalizers is assumed, the degrees of freedom after imposing the zero-interference constraint are increased from 2N T − N R to 2N T − S. The downside is that ISI is removed but ICI is not.…”

“…However, satisfactory results are only guaranteed if S = N R ≤ N T , where S is the number of streams spatially multiplexed on each subcarrier and N T , N R are respectively the number of transmit and receive antennas. Adopting the system model of [4], the iterative algorithm presented in [5] improves the results when S < min (N R , N T ). Unfortunately the processing developed in [5] does not mitigate ICI, which yields an error floor in some multiantenna configurations.…”

“…Adopting the system model of [4], the iterative algorithm presented in [5] improves the results when S < min (N R , N T ). Unfortunately the processing developed in [5] does not mitigate ICI, which yields an error floor in some multiantenna configurations. The technique proposed in [6] improves the resilience against the channel frequency selectivity with respect to [4], [5], since no assumptions are made about the flatness of the channel.…”

“…Unfortunately the processing developed in [5] does not mitigate ICI, which yields an error floor in some multiantenna configurations. The technique proposed in [6] improves the resilience against the channel frequency selectivity with respect to [4], [5], since no assumptions are made about the flatness of the channel. Nevertheless, the number of variables required to represent the system model increases, which has an impact on the complexity.…”