Relay Hybrid Precoding Design in Millimeter-Wave Massive MIMO Systems

Abstract

This paper investigates the relay hybrid precoding design in millimeter-wave (mmWave) massive MIMO systems. The optimal design of the relay hybrid precoding is highly nonconvex, due to the six-order polynomial objective function, sixorder polynomial constraint, and constant-modulus constraints. To efficiently solve this challenging non-convex problem, we first reformulate it into three quadratic subproblems, where one of the subproblems is convex and the other two are non-convex. Then, we propose an iterative successive approximation (ISA) algorithm to attain the high-approximate optimal solution to the original problem. Specifically, in the proposed ISA algorithm, we first convert the two non-convex subproblems to convex ones by the relaxation of the constant-modulus constraints, and then we solve the three corresponding convex subproblems iteratively. We theoretically prove that the ISA algorithm converges to a Karush-Kuhn-Tucker (KKT) point of the original problem. Simulation results demonstrate that the proposed ISA algorithm achieves good performance in terms of achievable rate in both fullconnected and sub-connected relay hybrid precoding systems.

Performance Analysis of Multi-Cell Millimeter-Wave Massive MIMO Networks With Low-Precision ADCs

“…In [22] and [23], hybrid beamforming has been investigated in mmWave MIMO networks. A joint beam selection scheme for analog precoding has been proposed in [24] and a relay hybrid precoding design has been studied in [25]. Different from conventional wireless channels in cellular networks, spatial sparsity emerges as a dominant nature in mmWave propagations [26].…”

mmWave Amplify-and-Forward MIMO Relay Networks With Hybrid Precoding/Combining Design

“…4, we equally set the power of source node and the relay node, all to be N s . We compare our algorithm with the ADMM in [13], the ISA in [14] and the OMP in [10] in terms As expected, when the number of antennas at the relay node increases, the performance of all different algorithms improves because of the additional antenna gain. Our proposed method has the best achievable rate performance among the four methods except for N r = 48.…”

“…So, for each outer iteration, the total number of inner iterations is O(log 2 1 ε ) + O(log 2 2 1 ε ). Compared with the algorithm in [14], for which the number of inner iterations is O(2(2N RF N r ) 2.5 log 1 ε ) for each outer iteration, the complexity of our algorithm is much lower especially for large antenna arrays.…”

“…In this section, we consider a relay MIMO system consisting of one source node equipped with a N t = 64 antenna array, a relay node with an N r = 32 antenna array and a destination node with a N d = 48 antenna array unless other number of antennas are specifically mentioned. The number of antennas is chosen from [14] for the purpose of the comparison. For simplicity, we use the channel model in Eq.…”

“…For sub-connected structures, [14] proposes a MMSE-based relay hybrid precoding design. To make the problem tractable, [14] reformulates the original problem as three subproblems and proposes an iterative successive approximation (ISA) algorithm. The algorithm in [14] can also be extended to the fully-connected structure.…”