We propose the joint dynamic power allocation and multi-relay selection for the cohabitation of high-priority military radar and low-priority commercial 5G communication. To improve the 5G network performance, we design the full-duplex underlay cognitive radio network for the low-priority commercial 5G network, where multiple relays are selected for concurrently receive the signal from the source and send it to the destination. Then, we propose the interference suppression at the high-priority radar system by using both non-coherent and coherent relay cases. In particular, we formulate the optimization problem for maximizing the system rate, with the consideration of the power constraints at the 5G users and the interference constraint at the radar system. Then, we propose the mathematical analysis model to evaluate the rate performance, considering the impacts of self-interference at the relays and derive the algorithms of joint power allocation and relay selection. Our numerical results demonstrate the characteristic of the optimal configuration and the significant performance gain of coherent case with respect to the non-coherent case and the existing algorithms with single relay selections.

翻译：针对高优先级军用雷达与低优先级商用5G通信的共存场景，本文提出联合动态功率分配与多中继选择方案。为提升5G网络性能，我们为低优先级商用5G网络设计了全双工底层认知无线电网络，通过选择多个中继并行接收源节点信号并转发至目的节点。随后，我们分别基于非相干与相干中继场景，提出面向高优先级雷达系统的干扰抑制方法。具体而言，在考虑5G用户功率约束与雷达系统干扰约束的前提下，构建了以系统速率最大化为目标的优化问题。通过建立数学模型分析中继自干扰对速率性能的影响，推导出联合功率分配与中继选择算法。数值仿真结果揭示了最优配置的特性，并证明相干方案相较于非相干方案及现有单中继选择算法具有显著的性能增益。