| ID | 原文 | 译文 |
| 4673 | 同时，将属性标识作为流表匹配字段，通过属性标识定义网络转发行为，该机制与属性签名验证共同实现细粒度的访问控制。 | At the same time, attribute identification was defined as a match field of flow by the framework, and the network for-warding behavior was defined based on attribute identification. A fine-grained access control was implemented by the proposed mechanism and attribute-based signature. The proposed mechanism and attribute-based signature implementeda fine-grained access control. |
| 4674 | 实验结果表明，该系统能有效实现数据流的细粒度的转发认证，且转发粒度高于同类方案。 | Experimental results demonstrate that the method can effectively implement fine-grainedforwarding and flow authentication, and the forwarding granularity is higher than that of similar schemes. |
| 4675 | 针对 5G 超蜂窝网络的宏基站和微基站群之间的非实时业务传输，提出了一种新型预测性信道调度算法。 | A novel predictive channel scheduling algorithm was proposed for non-real-time traffic transmission betweenmacro-base stations and micro-base stations in 5G ultra-cellular networks. |
| 4676 | 首先，利用静止通信体之间无线信道的随机平稳过程特征，从古典概率论的角度为调度过程建立了一个离散信道状态概率空间并对其进行事件域分割； | First, based on the stochastic stationary process characteristics of wireless channels between stationary communication agents, a discrete channel state probability space was established for the scheduling process from the perspective of classical probability theory, and the event do-main was segmented. |
| 4677 | 然后，通过各事件域的概率数值计算实现对多用户多非实时业务的高效调度。 | Then, the efficient scheduling of multi-user, multi-non-real-time services was realized by probabil-ity numerical calculation of each event domain. |
| 4678 | 理论分析和仿真结果表明，所提算法计算复杂度低，和其他一些经典调度算法相比，所提算法可在更长时间维度上进行业务优化传输，在吞吐量性能上逼近最大信噪比算法，重负荷情况下系统吞吐量提升了约 14%； | The theoretical analysis and simulation results show that the algorithm has low computational complexity. Compared with other classical scheduling algorithms, the new algorithm can optimizetraffic transmission in a longer time dimension, approximate the maximum signal-to-noise ratio algorithm in throughputperformance, and increase system throughput by about 14% under heavy load. At the same time, the new algorithm isaccurate. |
| 4679 | 同时通过精确量化计算实现了业务期望速率和实际调度速率的自适应匹配。 | Quantitative computation achieves a self-adaption match between the expected traffic rate and the actual sche-duling rate. |
| 4680 | 为识别当前卫星通信系统所采用的主要调制方式，提出了一种基于多端卷积神经网络的通信信号调制识别算法。 | In order to identify the main modulation modes adopted in current satellite communication systems, a signalmodulation recognition algorithm based on multi-inputs convolution neural network was proposed. |
| 4681 | 利用信号的先验信息以及对网络拓扑结构的认知，将信号时域波形转化为眼图和矢量图，作为信号的浅层特征表达，并由此设计了基于多端卷积神经网络的调制识别模型。 | With the prior infor-mation of the signals and knowledge of the network topological structure, the time-domain signal waveforms were con-verted into eye diagrams and vector diagrams to represent the shallow features of the signals. Meanwhile, the modulation recognition model based on multi-inputs convolution neural network was designed. |
| 4682 | 通过训练所搭建的网络，对浅层特征进行深度提取和映射，最终完成了目标信号的调制识别。 | Through the training of the network,the shallow features were deeply extracted and mapped. |