Two-dimensional Hybrid Precoding with Limited Feedback for 3-D Massive MIMO

3-D massive multiple input multiple output (MIMO) is one of the promising technologies to address the growing capacity requirement in fifth-generation cellular systems. However, the 3-D massive MIMO brings new challenges on channel state information (CSI) feedback because of its high-dimension complex channel matrix. Based on hybrid precoding, a hybrid limit feedback algorithm is proposed in frequency-division duplex (FDD) systems. By exploiting Kronecker structure of the 3-D MIMO channel matrix, we decompose the radio-frequency (RF) codebook into elevation-domain and azimuth-domain codebooks respectively to reduce the dimension of the matrix. Hence, a two-dimensional hybrid feedback algorithm based on Kronecker product is proposed in this paper. By introducing the equivalent channel, we descend the diagonal elements of the equivalent channel and select codewords of RF precoder(combiner) in elevation and azimuth respectively. The simulation results show that comparing to other two feedback algorithms, the proposed algorithm could achieve similar performance with much lower complexity.