On the design of hybrid limited feedback for massive MIMO systems

The increase of antennas can greatly improve the performance of MIMO systems, in the meanwhile, the precise channel state information is difficult to obtain since the channel matrix of massive MIMO is much more complex than the traditional one. In this paper, we propose a limited feedback strategy named hybrid limited feedback with selective eigenvalue information (HLFSEI), which jointly considers the conventional quantized feedback and codebook based feedback. In HLFSEI, because of the large amount of the elements of channel matrix, the quantized information of only selective eigenvalue elements is fed back from the receiver to the transmitter with feedback-link capacity constraint. We study the optimal feedback bit allocation of both feedback methods to maximize the throughput by theoretic deduction. Finally, we evaluate the performance of the proposed limited feedback scheme by simulation and show its performance gain compared to conventional feedback strategies.