IRS-Assisted Wireless Powered IoT Network With Multiple Resource Blocks
IRS-Assisted Wireless Powered IoT Network With Multiple Resource Blocks
复制标题
DOI：
10.1109/tcomm.2023.3242365
复制
发表时间：
2023-04
影响因子：
8.3
通讯作者：
Zheng Chu;Pei Xiao;D. Mi;Wanming Hao;Qingchun Chen;Yue Xiao
中科院分区：
文献类型：
--
作者：
Zheng Chu;Pei Xiao;D. Mi;Wanming Hao;Qingchun Chen;Yue Xiao
在本文中，我们研究了一种智能反射面（IRS）辅助的无线供电物联网（WP-IoT）网络，该网络在多个资源块（RB）中运行。特别地，IRS以改进WET中从发电站（PS）到各种IoT设备的能量反射并且提升WIT中从IoT设备到接入点（AP）的信息递送的方式，帮助下行链路无线能量传输（WET）和上行链路无线信息传输（WIT）。这些IoT设备能够利用收集的能量，并且在上行链路WIT中采用时分多址（TDMA）或非正交多址（NOMA）方案。为了最大限度地提高平均吞吐量作为考虑网络的整体性能指标，我们共同优化的PS，时间调度和IRS相移的发射功率分配。这些耦合变量导致该优化问题的非凸性，不能直接求解。为了解决这个问题，我们首先为每个RB设计最佳PS的发射功率分配。对于基于TDMA的方案，我们设计了上行链路WIT的闭式IRS波束图。然后，通过拉格朗日对偶方法和Karush-Kuhn-Tucker（KKT）条件推导出了下行链路和上行链路时间分配的闭式解。此外，提出了基于二次变换（QT）的交替方向乘法器（ADMM）方法，以交替的方式迭代地推导出下行链路WET的次优IRS波束方向图。对于基于NOMA的方案，我们提出了一种交替优化（AO）算法来迭代优化IRS相移，其中上行IRS波束方向图采用黎曼流形优化（RMO）方法迭代设计，而下行IRS相移采用基于QT的ADMM方法来交替获得次优。最后，数值结果表明，所提出的解决方案相比基准计划的性能有所改善，也突出了IRS在多RB场景中的应用优势。
In this paper, we investigate an intelligent reflecting surface (IRS)-assisted wireless powered Internet of Things (WP-IoT) network that operates in multiple resource blocks (RBs). Particularly, the IRS helps in both downlink wireless energy transfer (WET) and uplink wireless information transfer (WIT), in a way that it improves energy reflection in WET from a power station (PS) to various IoT devices and boosts information delivery in WIT from the IoT devices to an access point (AP). Those IoT devices are capable of utilizing the collected energy, and adopting the time-division multiple access (TDMA) or non-orthogonal multiple access (NOMA) scheme in the uplink WIT. Aiming to maximize the average throughput as the overall performance indicator of the considered network, we jointly optimize the transmit power allocation of the PS, the time scheduling, and the IRS phase shifts. These coupled variables lead to the non-convexity of this optimization problem, which cannot be solved directly. To address this problem, we first design the optimal PS’s transmit power allocation for each RB. For the TDMA-based scheme, we design the closed-form IRS beam pattern of the uplink WIT. Then, the closed-form downlink and uplink time allocations are derived by the Lagrange dual method and the Karush-Kuhn-Tucker (KKT) conditions. In addition, the quadratic transformation (QT)-based Alternating Direction Method of Multipliers (ADMM) approach is proposed to iteratively derive the sub-optimal IRS beam pattern of the downlink WET in an alternated fashion. For the NOMA-based scheme, we propose to apply an alternating optimization (AO) algorithm to iteratively optimize the IRS phase shifts, where the uplink IRS beam pattern is iteratively designed by the Riemannian Manifold Optimization (RMO) approach, and the QT-based ADMM method is adopted to alternately derive the sub-optimal downlink IRS phase shifts. Finally, numerical results demonstrate the improved performance of the proposed solution approaches compared to the benchmark schemes, also highlight advantages of the application of IRS in multiple RB scenarios.