ID	原文	译文
4073	仿真实验表明，在非合作通信系统中能准确识别 OFDM 信号与单载波信号，能对接收方 OFDM 信号子载波中的空子载波、QPSK 和 16QAM 达到理想的调制识别效果，克服了信道传输带来的相移和频偏问题，提高了调制方式识别的准确率。	Simulation experiments show that in non-cooperative communication systems, OFDM signals and sin-gle-carrier signals can be accurately identified, and ideal modulation recognition effects can be achieved on empty sub-carriers, QPSK, and 16QAM in the receiver OFDM signal subcarriers, overcoming the channel transmission band the problems of phase shift and frequency offset have improved the accuracy of modulation mode identification.
4074	针对先进持久性威胁场景中模拟防御系统安全性能评估的不足，基于 FlipIt 博弈论模型，提出了一种改进的博弈模型。	To make up for the lack of security performance evaluation of the mimic defense systems in the advanced per-sistent threat scenarios an improved game model based on the FlipIt game theory model was proposed.
4075	对不同的异构性条件下的拟态防御动态策略进行评估，并设计案例进行仿真分析。	The dynamicstrategy of mimic defense under different heterogeneity conditions was evaluated, and a case study for the simulationanalysis was conducted.
4076	仿真结果表明，不定周期的轮换能够弥补异构性的不足，维持防御者较高的博弈收益。	The simulation results show that the rotation of indefinite period can make up for the lack of he-terogeneity and maintain the higher game payoff of defenders.
4077	针对三维点云语义分割中缺乏结合点云的上下文细粒度信息导致的欠分割问题，提出一种基于上下文注意力卷积神经网络的三维点云语义分割算法。	Aiming at the under-segmentation of 3D point cloud semantic segmentation caused by the lack of contextualfine-grained information of the point cloud, an algorithm based on contextual attention CNN was proposed for 3D pointcloud semantic segmentation.
4078	首先，通过注意力编码机制挖掘点云的局部区域内细粒度特征；	Firstly, the fine-grained features in local area of the point cloud were mined through the at-tention coding mechanism.
4079	然后，通过上下文循环神经网络编码机制捕捉多尺度局部区域之间的上下文特征，且与细粒度局部特征相互补偿；	Secondly, the contextual features between multi-scale local areas were captured by the con-textual recurrent neural network coding mechanism and compensated with the fine-grained local features.
4080	最后，采用多头部机制增强网络的泛化能力。	Finally, themulti-head mechanism was used to enhance the generalization ability of the network.
4081	实验结果表明，所提算法在 ShapeNet Parts、S3DIS 和 vKITTI 标准数据集上的平均交并比分别为 85.4%、56.7%和 38.1%，分割性能良好，且具有较好的泛化能力。	Experiments show that the mIoU of the proposed algorithm on the three standard datasets of ShapeNet Parts, S3DIS and vKITTI are 85.4%, 56.7% and38.1% respectively, which has good segmentation performance and good generalization ability.
4082	针对大型网络中效率计算时间复杂度高、计算耗时长的问题，提出一种基于分形特性的网络效率估计方法。	A method of network efficiency estimation based on fractal property was proposed for solving the problems ofhigh complexity and time-consuming calculation of efficiency in large scale networks.