ID	原文	译文
53687	研究和仿真结果表明: 在信道具有受限多普勒功率谱时，改进拼接法的自相关函数的均方根误差明显好于另外两种方法，	The results show that the root mean square error of the autocorrelation function of the improved splicing method is significantly lower than the other two methods when considering a restricted Doppler power spectrum.
53688	同时还能避免设计数字滤波器和希尔伯特变换器，但是改进拼接法的算法复杂度远远大于其他两种模型，复杂度最小的是单路滤波法，所以在实际仿真中需要看情况取舍。	Meanwhile, the digital filters and Hilbert converters can be left out. The single-path filtering method leads to the lowest complexity cost.
53689	论文所做比较有利于受限多普勒谱建模工作的模型选取，并为更加复杂的多普勒功率谱建模提供参考。	The contributions in the paper can help to select the constrained Doppler spectrum model and provide a reference for more complicated Doppler power spectrum modeling.
53690	针对骨架行为识别对时空特征提取不充分以及难以捕捉全局上下文信息的问题，研究了一种将时空注意力机制和自适应图卷积网络相结合的人体骨架行为识别方案。	To solve the problem that skeleton behavior recognition can not extract spatio-temporal features sufficiently and it is difficult to capture global context information, a human skeleton behavior recognition scheme based on spatio-temporal attention mechanism and adaptive graph convolution network is studied.
53691	首先，构建基于非局部操作的时空注意力模块，辅助模型关注骨架序列中最具判别性的帧和区域;	Firstly, a spatio-temporal attention module based on non-local operation is constructed to assist the model to focus on the most discriminative frames and regions in the skele- ton sequence;
53692	其次，利用高斯嵌入函数和轻量级卷积神经网络的特征学习能力，并考虑人体先验知识在不同时期的影响，构建自适应图卷积网络;	secondly, an adaptive graph convolution network is constructed by using the feature learning ability of Gaussian embedding function and lightweight convolution neural network, and considering the effect of human prior knowledge in different time periods;
53693	最后，将自适应图卷积网络作为基本框架，并嵌入时空注意力模块，与关节信息、骨骼信息以及各自的运动信息构建双流融合模型。	finally, the adaptive graph convolution network is used as the basic framework, the spatiotemporal attention module is embedded to construct two-stream fusion model with joint information, bone information and their respective motion information.
53694	该算法在 NTU ＲGB+D 数据集的两种评价标准下分别达到了 90. 2% 和 96. 2% 的准确率，	The accuracy of the algorithm is 90. 2% and 96. 2% respectively under the two evaluation standards of NTU ＲGB+D dataset.
53695	在大规模的数据集 Kinetics 上体现出模型的通用性，验证了该算法在提取时空特征和捕捉全局上下文信息上的优越性。	The universality of the model is reflected in the large-scale dataset Kinetics, which verifies that the algorithm is proved to be superior in extracting spatio-temporal features and capturing global context information.
53696	基于稀疏恢复的空时自适应算法( Sparse Ｒecovery Space Time Adaptive Processing，SＲ-STAP) 能有效改善机载雷达在复杂环境下对杂波的抑制能力，	The Sparse Ｒecovery Space Time Adaptive Processing ( SＲ-STAP) algorithm can effectively improve the cluttersuppression performance of airborne radar in complex environment.