Dynamic Small Cell Clustering and Stackelberg Game-Based Precoding Design for Two-Tier Heterogeneous Networks with Massive MIMO

In this paper, we investigate the dynamic small cell (SC) base station (BS) clustering strategy and their precoding design problem for interference coordination in two-tier heterogeneous networks (HetNets) with massive MIMO (mMIMO). First, to reduce interference among different SCs, an interference graph-based dynamic SC BS (SBS) clustering scheme is proposed. Then, we formulate an optimization problem as design precoding weights at MBS and clustered SBSs for maximizing the downlink sum rate of SUs subject to per-SBS power constraint, while mitigating the inter-tier, inter-cluster, intra-cluster and multimacro user (MU) interference. To eliminate the inter-tier and multi-MU interference simultaneously, we propose a clustered SC block diagonalization (CSBD) precoding scheme for macro BS (MBS). Next, the precoding vector of each SC user (SU) at clustered SBSs is designed as the product of the following two matrices. The first matrix is designed with singular value decomposition (SVD) to remove the intra-cluster interference. The second matrix is designed to coordinate the inter-cluster interference while maximizing the downlink sum rate of SUs, which is a non-convex optimization problem, and we transform it into a Stackelberg game model. Then, we prove the existence and uniqueness of Nash Equilibrium (NE) for the formed game and propose an iterative algorithm to obtain the NE. Finally, simulation results verify the effectiveness of our proposed schemes.