ID 原文 译文
20155 同时由于不存在流形映射或者模式截断误差，估计结果具有统计无偏性。 At the same time, the results are statistically unbiased because no manifold mapping or mode truncation error is introduced.
20156 仿真结果验证了所提算法的有效性。 Simulations verify the effectiveness of the proposed algorithm.
20157 针对经典的快速区域卷积神经网络(Faster RCNN)训练过程存在太多难训练样本、召回率低等问题，该文采用一种基于在线难分样本挖掘技术(OHEM)与负难分样本挖掘(HNEM)技术相结合的方法，通过训练中实时筛选的最大损失值难分样本进行误差传递，解决了模型对难分样本检测率低问题，提高模型训练效率； Because of the classic Faster RCNN training proccess with too many difficult training samples andlow recall rate problem, a method which combines the techniques of Online Hard Example Mining (OHEM) and Hard Negative Example Mining (HNEM) is adopted, which carries out the error transfer for the difficult samples using its corresponding maximum loss value from real-time filtering. It solves the problem of low detection of hard example and improves the efficiency of the model training.
20158 为更好地提高模型的召回率和模型的泛化性，该文改进了非极大值抑制(NMS)算法，设置了置信度阈值罚函数，又引入多尺度、数据增强等训练方法。 To improve the recall rate and generalization of the model, an improved Non-Maximum Suppression (NMS) algorithm is proposed by setting confidence thresholds penalty function; In addition, multi-scale training and data augmentation are also introduced.
20159 最后通过比较改进前后的结果，经敏感性实验分析表明，该算法在VOC2007数据集上取得了较好效果，平均精度均值从69.9%提升到了74.40%，在VOC2012上从70.4%提升到79.3%，验证了该算法的优越性。 Finally, the results before and after improvement are compared: Sensibility experiments show that the algorithm achieves good results in VOC2007 data set and VOC2012 data set, with the mean Average Percision (mAP) increasing from 69.9% to 74.40%, and 70.4% to 79.3% respectively, which demonstrates strongly the superiority of the algorithm.
20160 该文考虑一个大规模MIMO全双工中继系统，该系统下多个单天线的源节点和多个单天线的目的节点通过一个在接收端和发送端分别配置和根天线的中继进行通信。 A massive MIMO full-duplex relaying system is considerd in this paper, in which multiple single-antenna sources simultaneously communicate with multiple single-antenna destinations using a single relay that is equipped with  receive antennas and  transmit antennas.
20161 在非完美信道状态信息(CSI)和硬件损害的情况下，中继端在译码转发(DF)的方式下使用迫零(ZF)方法去处理接收信号和发送信号，并获得了闭合形式的接收速率表达式。 Under imperfect Channel State Information(CSI) and hardware impairment, the relay processes the received and transmitted signals by means of Zero-Forcing (ZF) and uses Decode-and-Forward (DF) scheme. The closed-form expressions of achievable rate are deduced.
20162 根据这些表达式，可以得到各种功率收缩规律，发现随着中继端接收天线数量和发送天线数量以一定比例趋近于无穷大时，源节点的能量、中继端的能量和导频的能量可以以一定条件收缩来维持系统性能。 Based on these expressions, the various power scaling laws can be obtained. It is shown that when thetwo numbers of the relay receive and transmit antennas go to infinity but with a fixed ratio, the system canmaintain a desirable quality of service in the case of scaling the transmit powers of the sources, relay and pilots.
20163 针对目前民航运输业对航班延误高精度预测的需求，该文提出一种基于深度SE-DenseNet的航班延误预测模型。 Nowadays, the civil aviation industry has a high-precision prediction demand of flight delays, thus a flight delay prediction model based on the deep SE-DenseNet is proposed.
20164 该模型首先将航班信息、相关机场延误信息和天气信息进行数据融合； Firstly, flight data, associated airport delay information and meteorological data are fused in the model.