| ID | 原文 | 译文 |
| 6394 | 相比传统算法,该方法具有更好的抗低信噪比和抗混叠干扰能力,验证了卷积神经网络在特定信号识别领域的有效性,为该领域的后续研究奠定了基础。 | Compared with traditional algorithm, this method has better resistance to low signal-to-noise ratio and the ability to anti aliasing, verify the convolutional neural network in the field of specific signal recognition effectiveness, which laid a foundation for further research in this field. |
| 6395 | 针对激光雷达低空风切变信号图像的类型识别问题,提出了一种基于深度卷积神经网络(deep convolutional neural network,DCNN)的多层特征提取及自适应融合算法。 | For laser radar low-level wind shear of signal type recognition problem, this paper proposes a convolutional neural network based on depth (deep convolutional neural network, the DCNN) multilayer feature extraction and adaptive fusion algorithm. |
| 6396 | 该方法可以有效解决网络逐层训练过程中信息丢失的问题。 | The method can effectively solve the problem of information loss in the process of network training step by step. |
| 6397 | 首先,采用DCNN提取低空风切变信号图像的各层网络特征,并将各特征进行L2范数标准化实现同趋化。 | First of all, the low-level wind shear signal extracted with the use of DCNN image characteristics of each layer of network, and the characteristics of L2 norm standardization implementation with chemokines. |
| 6398 | 其次,将其以多通道图像形式输入单层CNN进行自适应融合,将融合特征送入支持向量机进行分类识别。 | Second, to be in the form of a multi-channel image input single CNN for adaptive fusion, the fusion feature into the support vector machine (SVM) classification recognition. |
| 6399 | 结果表明,采用所提算法进行低空风切变图像类型识别的平均识别率为98.1%,与其他4种算法相比均有提升。 | The results show that the proposed algorithm is used to identify the image type low-level wind shear of the average recognition rate was 98.1%, compared with other four kinds of algorithms are improved. |
| 6400 | 所提算法能更有效地实现低空风切变信号图像类型识别。 | The proposed algorithm can more effectively realize the low-level wind shear signal image type identification. |
| 6401 | 在分析微弱信号条件下残余多普勒对全球导航卫星系统(global navigation satellite system,GNSS)信号捕获影响的基础上,提出了基于迫零(zero forcing,ZF)频率误差修正的GNSS信号捕获算法。 | On the analysis of the weak signal under the condition of residual doppler on global navigation satellite system (global navigation satellite system, GNSS) signal capture effect, on the basis of forced is proposed based on zero (zero forcing, ZF) frequency error correction algorithm of GNSS signal capture. |
| 6402 | 该算法将ZF的频率修正与快速改进的双块零扩展(fast modified double block zero padding,FMDBZP)算法有机结合,从而获得高精度的频率估计结果,并将修正项实时引入本地环路振荡器,减少了由多普勒频率误差引起的相关功率损失。 | The algorithm will ZF frequency correction and rapid improvement of double block zero extension (fast modified double block zero padding, FMDBZP) organic combination algorithm, to obtain high accuracy of frequency estimation results, and will introduce a correction term real-time local loop oscillator, reduce the error caused by doppler frequency related power losses. |
| 6403 | 理论分析和仿真实验表明,在给定仿真条件下,该算法能明显提高捕获性能,捕获灵敏度相对于FMDBZP算法有1.2dB的性能提升,残余多普勒频率估计的标准差小于2Hz。 | The theoretical analysis and simulation experiments show that under the condition of a given simulation, this algorithm can significantly improve the performance of capture capture sensitivity relative to FMDBZP algorithm is 1.2 dB performance improvement, residual doppler frequency estimation of the standard deviation is less than 2 hz. |