智能反射面支持的空中基站在下行NOMA异构网络中的功耗最小化与干扰抑制

Intelligent reflection surface (IRS) is an advanced technology that effectively transforms the wireless propagation environment to next-generation wireless networks. In this paper, we employ the IRS-supported multi-aerial base station (ABS) in a downlink non-orthogonal multiple access (NOMA) heterogeneous network (HetNets), where the transmission signals from multiple ABSs to ground users (GUs) are enhanced using the IRS. Our goal is to maximize the system sum rate (SSR) by mitigating inter-cluster and intra-cluster interference and to maximize system energy efficiency (SEE) by minimizing ABS power transmission while preserving the required minimum data rate. The formulated optimization problem is non-convex due to the joint optimization of ABS power transmission, ABS three-dimensional (3D) positions, IRS reflection angles (RAs), IRS reflection coefficients (RCs), and decoding order (DEO) among GUs. To tackle this problem, a meta-heuristic algorithm based on the modified gray wolf optimization (MGWO) and an alternation max/min technique based on the developed Block Coordinate Descent (DBCD) are proposed. In particular, the original optimization problem is divided into four sub-problems; the first three sub-problems (RS RAs, IRS RCs, and ABS 3D positions) are alternately addressed using the MGWO technique; and the fourth sub-problem (ABS power transmission) is addressed using the DBCD technique. Finally, a dynamic DEO strategy is utilized to address the NOMA DEO among GUs. The numerical results demonstrated that the proposed schemes outperformed the conventional schemes, achieving up to a 43.289% improvement in SSR and a 71.536% enhancement in SEE, while minimizing the total power consumption. Moreover, the results demonstrate that integrating IRS into multi-ABS HetNets significantly improves the total performance by enhancing the quality of the channel between ABSs and their corresponding GUs while minimizing inter-ABS interference. Finally, by strategically optimizing ABS 3D positions and utilizing DEO among GUs, the SSR and SEE gain achieved by NOMA is shown to be more significant compared to conventional methods.