Downlink Multi-User Scheduling With Zero-Forcing Precoding in Cognitive Hetnets: From Performance Tradeoff Perspective

In a cognitive heterogeneous network (HetNet), cognitive small cell base stations (SBSs) can perceive and utilize the frequency band belonging to massive MIMO enabled macrocell base stations opportunistically. There is a performance tradeoff between the macrocell-tier and the SC-tier. To restrict the cross-tier interference from SBSs to macrocell users (MUs), we adopt the concept of exclusion region (ER). A multi-user scheduling with zero-forcing precoding strategy is proposed to maximize the utility of the entire HetNet. The involved multi-user scheduling problem is formulated as the maximization of a general utility function. However, the problem is intractable for two reasons: First, it depends on ergodic rate bothering online scheduling. Second, ERs formed by different scheduled MUs may intersect with each other which makes the problem difficult to solve. Therefore, we convert the maximization problem into a simplified version and design an interfering SBS cluster splitting-based iterative user selection algorithm to figure out the scheduled MU set and its corresponding precoder. Simulation results reveal that the proposed scheme achieves significant gains compared with existing strategies and that the proposed strategy is efficient for dense HetNets.