ID	原文	译文
3463	该方法提取数字信号的高阶累积量作为信号特征，综合利用判别式受限玻尔兹曼机的生成能力和分类能力，分析了含有高斯噪声、时变相位偏移或瑞利衰落环境下的数字信号识别率。	which extracted the high-order cumulant of digital signals as signal features, comprehensively uti-lized the generation ability and classification ability of the discriminative restricted Boltzmann machine, analyzed therecognition rate of digital signals in environments containing Gaussian noise, time-varying phase offset or Rayleigh fad-ing.
3464	实验结果表明，与传统识别方法相比，所提方法的识别性能有明显改善。	Experimental results show that compared with traditional classification methods, the recognition performance of the proposed method is obviously improved.
3465	此外，利用该模型的生成能力对输入特征进行重构，可有效提高低信噪比下的信号识别率。	In addition, the use of the model's generation ability to reconstruct the input features can effectively improve the signal recognition rate under low signal-to-noise ratio.
3466	针对基于传统卷积神经网络模型的高光谱图像分类算法细节表现力不强及网络结构过于复杂的问题，设计了一种基于多尺度近端特征拼接网络的高光谱图像分类方法。	Aiming at the phenomenon that the hyperspectral classification algorithm based on traditional CNN model wasnot expressive enough in detail and the network structure was too complex, a hyperspectral image classification methodbased on multi-scale proximal feature concatenate network (MPFCN) was designed.
3467	通过引入多尺度滤波器和空洞卷积，在保持模型轻量化的同时可以获取更丰富的空间−光谱判别特征，并提出利用卷积神经网络近端特征间的相互联系进一步增强细节表现力。	By introducing multi-scale filter andcavity convolution, the model could be kept light and the discriminative features of the space spectrum could be obtained,and the correlation between the proximal features of the CNN was proposed to further enhance the detail expression.
3468	在 3 个基准高光谱图像数据集上的实验结果表明，所提方法优于其他分类模型。	Ex-perimental results on three benchmark hyperspectral image data sets show that the proposed method is superior to other classification models.
3469	针对通信辐射源个体识别问题，提出了一种基于多通道变换投影、集成深度学习和生成对抗网络的融合分类方法。	A multi-feature fusion classification method based on multi-channel transform projection, integrated deeplearning and generative adversarial network (GAN) was proposed for communication specific emitter identification.
3470	首先，通过对原始信号进行多种变换得到三维特征图像，据此构建信号的时频域投影以构建特征数据集，并使用生成对抗网络对数据集进行扩充。	First,three-dimensional feature images were obtained by performing various transformations, the time and frequency domainprojection of the signal was constructed to construct the feature datasets. GAN was used to expand the datasets.
3471	然后，设计了一种基于多特征融合的双阶段识别分类方法，利用神经网络初级分类器分别对 3 类特征数据集进行学习，得到初始分类结果。	Then, a two-stage recognition and classification method based on multi-feature fusion was designed. Deep neural networks wereused to learn the three feature datasets, and the initial classification results were obtained.
3472	最后，通过叠加融合学习初始分类结果，得到最终的分类结果。	Finally, through fusion andre-learning of the initial classification result, the final classification result was obtained.