ID	原文	译文
3883	通过在无人机以及用户设备上部署存储设备，将用户请求的热点内容缓存后进行共享分发，以减少无人机回程链路资源与能量开销，减轻业务负载并降低用户时延。	By de-ploying caches on UAV and user device, the popular content requested by user was cached and delivered, which alleviat-ed the backhaul resource and energy consumption of UAV, reduced the traffic load and user delay.
3884	以用户内容获取时延最小化为目标建立了无人机与用户缓存联合优化问题，将其分解为无人机缓存子问题与用户缓存子问题，并分别基于交替方向乘子法和全局贪婪算法进行求解，通过迭代获得收敛的优化结果，实现了无人机与用户的协同缓存。	A joint optimization problem of UAV and user caching was established with the goal of minimizing user content acquisition delay, and de-composed into UAV caching sub-problem and user caching sub-problem, which were solved based on alternating direc-tion method of multiplier and global greedy algorithm respectively. The iterative way was used to obtain convergent op-timization result, and the cooperative caching of UAV and user was realized.
3885	仿真结果表明，所提算法能够有效降低用户的内容获取时延，提升系统性能。	Simulation results show that the proposedalgorithm can effectively reduce user content acquisition delay and improve system performance.
3886	为了缓解频谱资源匮乏、提高认知无线网络的功率利用率，提出了一种基于能量收集的全双工认知中继网络功率分配算法。	To alleviate the shortage of spectrum resources and improve the power utilization of cognitive radio networks,a resource allocation algorithm of full duplex cognitive relay networks with energy harvesting was proposed.
3887	所提算法在保证对主用户干扰以及中继端收集能量满足一定条件的情况下，优化能量受限的认知中继节点功率分配系数及次用户发射功率，实现了次用户系统吞吐量的最大化，提高了频谱效率。	In the algo-rithm, the coefficient for power splitting of the relay and the transmit power of the secondary users were jointly optimizedto maximize the throughput of the secondary users under the interference to primary users and energy harvesting con-straints.
3888	由于所提算法的优化问题是非凸的，因此将其转化为针对功率分配系数和次用户发射功率的 2 个可解凸优化子问题，通过收敛迭代算法求得原问题的最终解。	Since the optimization of the algorithm was non-convex, it was transformed into two sub-optimizations, thesub-optimization of the coefficient for power splitting and the sub-optimization of the power transmitted of secondary us-ers, which were the solvable convex sub-optimizations.
3889	仿真结果表明，采用所提算法的次用户系统吞吐量，与采用半双工功分算法的系统吞吐量相比提升了一倍，与采用全双工时分算法的系统吞吐量相比提升了 0.5 倍。	Then, the final solution of the original optimization was obtainedwith the iterative algorithm. Simulation results show that the throughput of the proposed algorithm, can obtain 2 timesthroughput of the networks with half-duplex power splitting algorithm and 1.5 times throughput of the networks withfull-duplex time switching algorithm.
3890	考虑到用户与基站间的相关性和网络的不规则性，建立了基于 Matern 簇过程（MCP）的非正交多址接入型异构网（NOMA-HetNet）模型，并结合该模型分析了各类用户的覆盖概率。	Considering the correlation between the users and base stations and the network irregularity, thenon-orthogonal multiple access heterogeneous networks (NOMA-HetNet) based on Matern cluster process (MCP) wasmodeled. The coverage probability of UEs was analyzed.
3891	在分析过程中，首先，从理论上推导出用户的信干噪比（SINR）；其次，针对现实中的非完美串行干扰消除（SIC）方案，给出了一种更符合实际情况的 SINR 上限更新规则；	Firstly, the signal-to-interference-plus-noise ratios (SINR) ofUEs were theoretically derived. Secondly, an updating law for the upper limit of SINR was presented for the imperfectserial interference elimination (SIC) scheme, which was more consistent with the reality.
3892	然后，利用空间坐标系、概率论、随机几何理论和顺序统计量等数学工具分析了各类用户关联概率和服务距离；	Then, using the mathematicaltools such as spatial coordinate system, probability theory, stochastic geometry theory and order statistics, the correlationprobability and service distances of the different types of UEs were analyzed.