Blind iterative singular vectors estimation and adaptive spatial loading in a reciprocal MIMO channel

In this paper spatial loading in a reciprocal time-varying multiple-input multiple-output (MIMO) channel is considered. We take inspiration from a blind iterative algorithm proposed in the literature to estimate the singular vectors associated to the dominant singular value to perform beamforming. Starting from that, we introduce an iterative algorithm that can estimate all the singular vectors and the associated singular values of the channel matrix. Then the estimated singular vectors are used to transmit over the parallel sub-channels and the associated singular values are considered to implement the rate and power loading algorithm described in this paper. A procedure based on the joint use of the two considered algorithms can be adopted to adaptively maximize the total rate for a given error rate performance and a given constraint on the average transmitted power. Simulation results are used to demonstrate the effectiveness of our approach compared to blind iterative beamforming transmission.