论文标题
部分可观测时空混沌系统的无模型预测
Energy-Efficient Power Control and Beamforming for Reconfigurable Intelligent Surface-Aided Uplink IoT Networks
论文作者
论文摘要
最近，可重新配置的智能表面（RIS）是由大量低成本反射元素组成的平面跨表面，由于它能够通过重新配置无线传播环境来提高光谱和能量效率的能力，因此受到了很大的关注。在本文中，我们提出了RIS相移和BS波束形成优化技术，该技术可以最大程度地减少RIS AID的IoT网络的上行链路传输功率。所提出的方案的关键思想，即基于Riemannian结合梯度的关节优化（RCG-JO），是使用Riemannian共轭梯度技术共同优化RIS相位移位和BS波束成形向量。通过利用单位模数相移和单位 - 摩尔型边缘向量的riemannian歧管结构，我们将NonConvex上行链路最小化问题转换为无约束的问题，然后在产品Riemannian歧管上找到最佳解决方案。从绩效分析和数值评估中，我们证明了所提出的RCG-JO技术可实现$ 94 \％$ $减少上行链路在没有RIS的情况下传输功率。
Recently, reconfigurable intelligent surface (RIS), a planar metasurface consisting of a large number of low-cost reflecting elements, has received much attention due to its ability to improve both the spectrum and energy efficiencies by reconfiguring the wireless propagation environment. In this paper, we propose a RIS phase shift and BS beamforming optimization technique that minimizes the uplink transmit power of a RIS-aided IoT network. Key idea of the proposed scheme, referred to as Riemannian conjugate gradient-based joint optimization (RCG-JO), is to jointly optimize the RIS phase shifts and the BS beamforming vectors using the Riemannian conjugate gradient technique. By exploiting the product Riemannian manifold structure of the sets of unit-modulus phase shifts and unit-norm beamforming vectors, we convert the nonconvex uplink power minimization problem into the unconstrained problem and then find out the optimal solution on the product Riemannian manifold. From the performance analysis and numerical evaluations, we demonstrate that the proposed RCG-JO technique achieves $94\%$ reduction of the uplink transmit power over the conventional scheme without RIS.
