ID	原文	译文
4753	仿真实验结果证明，所提算法具有较好的不可感知性，并且对常规信号处理（MP3 压缩等）和去同步攻击（音调伸缩等）均具有较好的稳健性。	The experimental results show that the proposed approach has not only good trans-parency, but also has strong robustness against common audio processing such as MP3 compression and good robustness against the desynchronization attacks such as pitch-scale modification et al.
4754	利用描述能力强且性能稳定的平滑梯度刻画局部数字音频性质，提出一种基于平滑梯度的平稳小波域音频特征点提取方法，有效解决音频特征点稳定性差且分布极不均匀的缺点，提高了数字音频水印对幅度伸缩、局部变调、随机剪切等去同步攻击的抵抗能力。	A SWT domain audio feature point extractionmethod based on smooth gradient is proposed, which effectively solves the drawbacks of poor stability and uneven dis-tribution of audio feature points, and improves the resistance of digital audio watermarks to amplitude-scale modification,pitch-scale modification, random cropping, and jittering attacks.
4755	在线社交网络的迅速发展使信息呈现爆炸式增长，然而不同消息的流行度存在较大差异，对其准确预测一直是领域内的研究难点。	The rapid development of online social networks leads to an explosion of information, however, there are greatdifferences in the popularity of different messages, and accurate prediction is always a great difficulty is the current study.
4756	流行度预测的任务是根据消息传播早期过程中涌现的特征预测其未来的传播趋势，	Popularity prediction of online content aims to predict the popularity in the future based on its early diffusion status.
4757	现有基于传播网络特征与拟合函数的预测模型难以解决预测准确率低的问题，因此借助社会学中的弱连接理论，引入连接强度的概念，并融合消息传播早期的流行度构建多元线性回归方程，提出了一种针对 Facebook 知名主页的消息流行度的预测模型 TSL。	Ex-isting models for popularity prediction were mostly based on discovering network features or fitting the equation into avarying time function that the accuracy of current popularity prediction model was not high enough. Therefore, with the help of the weak ties theory in sociology, the concept of tie strength was introduced and a multilinear regression equationwas constructed combined with the early popularity. A TSL model to predict the popularity of Facebook's well-knownpages was proposed. The main contribution of this article was to solve the problem and few or no work based on sociolo-gy. A high linear correlation between the proportion of faithful fans was existed in Facebook homepage with frequentshares in the early and the future popularity.
4758	通过在 Facebook 真实数据集（含 154 万次转发）上与其他具有代表性的基准模型进行比较，实验表明 TSL 模型可以对消息的最终转发流行度进行有效预测，预测性能优于同类方法。	Compared with other baseline models, an experimental study of Facebook(including 1.54 million shares) illustrates the effectiveness of the proposed TSL model, and the performance is better thanthe existing similar methods.
4759	针对蒸发波导时空态势感知过程中，压缩感知观测阶段混入噪声对高精度重构产生不利影响的问题，提出了基于二次最优阈值的近似消息传递法。	To solve the problem that the mixed noise of compressed sensing observation stage had adverse effects on thehigh-precision reconstruction in the time-space situation awareness process of evaporative duct, an approximate messagepassing method based on quadratic optimal threshold was proposed.
4760	所提方法综合了软、硬阈值函数的优势，设计了一种二次稳固阈值函数，改善 AMP 算法的重构性能，并在近似消息传递法的迭代过程中加入了基于共轭梯度法的最优阈值设置策略，加快收敛速度。	A quadratic stabilization threshold function whichintegrated advantages of soft and hard threshold functions was designed, and the optimal threshold setting policy wasadded to the iterative process based on conjugate gradient method to speed up convergence.
4761	理论分析和实验表明，所提算法比现有的几种主流重构算法的收敛速度更快，去噪重建误差更小。	Theoretical analysis and ex-periments show that, compared with several mainstream algorithms, the proposed algorithm has the faster convergencerate and lower denoising reconstruction error.
4762	针对跳频通信调制方式识别问题，提出了一种基于时频能量谱纹理特征的跳频调制方式识别方法。	For frequency hopping modulation identification, a novel method based on time-frequency energy spectrumtexture feature was proposed.