Low-Complexity Beam Selection Algorithms Based on SVD for mmWave Massive MIMO Systems

To realize beamspace MIMO with beam selection more efficiently, we propose low-complexity beam selection algorithms based on singular value decomposition (SVD). We first diagonalize the channel matrix by SVD, and select the appropriate beams one by one in a decremental or incremental order to maximize the sum-rate. To reduce the complexity of the proposed algorithms, we update eigenvalues by solving the secular equation instead of SVD from scratch again. Our method can achieve
$\mathcal {O}(N_{r}^{2}N_{c}^{2})$
, which is at most one in
$\frac {N_{t}+N_{r}}{N_{r}}$
of the state-of-the-art method. Analytical and simulation results demonstrate that our proposed algorithms outperform the state-of-the-art algorithms in terms of the sum-rate and complexity.