ID 原文 译文
12874 从用频装备单频和调幅连续波敏感度试验数据出发,基于用频装备带内电磁能量耦合及共性干扰、损伤作用机理分析，研究了有效值敏感和峰值敏感两类用频装备带内双频窄谱电磁辐射阻塞干扰效应预测模型并提出了预测方法，给出了受试用频装备适用预测模型的判定方法，并以某型超短波通信电台为受试对象，试验验证了模型及方法的有效性。 Based on the single frequency and the amplitude modulaton continuous wave electromagnetic radiation susceptibllity test data, considering the theoretical analysis of the electromagneic energy coupling and the interference damage acting mechansm for the electronic equipment, the dual-frequency in-band electromagnetic radiation blocking nterference effects predction models and predction methods for two kinds of electronc equipment are established separately. The one electronc equipment is sensitive to the effectve value of thterference signal. The other one is sensitive to the peak value. The decision method of choosing prediction model for the different electronic equipment is put forward. The typical ultra-short wave communication stations are selected as radio equpment under test. The test results verify the validity of the prediction model and the prediction method.
12875 研究结果表明：本文提出的模型和方法可有效应用于用频装备在带内双频电磁辐射作用下的阻塞干扰效应预测和评估。 The above researches indicate that the model and method presented in this paper can be effectively used to predict and evaluate the blocking interference effects for the radio equipment under the condition of the in-band dual-frequency electromagnetic radiation.
12876 为了分析东海蒸发波导高度的季节变化、月变化!讨论其分布规律，为雷达探测和无线电通讯提供参考，利用2008—2017年的十年NCEP-FNL再分析数据和改进的NPS蒸发波导模型，统计分析了我 国东海海域蒸发波导发生规律和时空分布特征。 Based on the final analysis data(FNL) from National Centers ；or Envronment Prediction (NCEP) reanalyss data and the improved NPS diagnostic model, the seasonal and monthly changes of evaporation duct over East Ch ina Sea are analyzed. The distribution of ducts is discussed to provide reference for radar detections and radio communications.
12877 结果表明：东海蒸发波导高度在春季和夏季呈西北高东南低的特点，其中春季为高度最高的季节5月份东海西北部可高达20 m，夏季6月份在东海中部出现最低值，秋季和冬季反之，呈西北低东南高的特点。 The results show that the evaporation duct height of northwest is higher than southeast in spring and summer. Spring is the season with the highest duct height. The maximum of evaporaton duct height over the northwest of East China Sea is in May, which is 20 m. For the central East Ch ina Sea, the lowest evaporaton duct height is in June. On the contrary, higher in southeast in autumn and winter.
12878 分析表明，东海蒸发波导高度存在明显的季节变化、月变化以及区域差异，这可能与东海的地理特征和气候变化相关。 There are obvious seasonal, monthly and regonal dfferences n the heght of the evaporaton duct over the East Chna Sea. Ths may be related to the geographcal characteristcs of the East China Sea and climate change.
12879 基于合成孔径雷达(synthetic aperture radar, SAR)在图像目标识别领域中识别精度低的问题，设计一种利用并联卷积神经网络(convolutional neural network, CNN)来提取SAR图像特征的目标识别方法。 In view of the low recognition accuracy of synthetc aperture radar(SAR)in the field of image target recognition, this paper designs a target recognition method for extracting SAR image features by shunt convolutonal neural network.
12880 首先利用改进的ELU激活函数代替常规的ReLU激活函数，建立与二次代价函数相结合的深度学习模型。 Firstly, the improved ELU activation function is used to replace the conventi onal ReLU act i vat i on funct i on, and a deep learni ng model comb i ned wi th the quadrat i c cost funcr ton is establshed.
12881 其次采用均方根支柱(root mean square Prop, RMSProp)与Nesterov动量结合的优化算法执行代 价函数参数迭代更新的任务，利用 Nesterov引入动量改变梯度，从两方面改进更新方式，有效地提高网络的收敛速度与精度。 Secondly, this paper uses the optimization algorithm combining root mean square prop (RMSProp) and Nesterov momentum to performthe teratve updatng task of cost functon parameters, and uses Nesterov to introduce momentum to change the gradent to improve the updating method from two aspects to effectvely improve the convergence speed and accuracy of the network.
12882 通过对美国国防研究规划局(DARPA)和空军研究实验室(AFRL)共同推出的MSTAR 数据集进行实验，实验表明，该文提出的算法能充分提取出SAR图像中各类目标所蕴含的信息，具有较好的识别性能，是ー种有效的目标识别算法。 Experiments on the MSTAR data set produced by DARPA and AFRL show that the proposed algorthm is effective in target recognton s nce that t can extract the nformaton contained in various targets in SAR mages sufficiently, and has good recognition performance.
12883 针对无线设备"指纹”特征提取技术含量较高，且技术手段较为复杂的问题,在无线空间信道状态不变的前提下，提出了一种基于卷积神经网络(convolution neural network, CNN)自动分类无线路由器 指纹的识别方法，解决无线设备"指纹”特征提取困难的问题。 Aiming at the problem of high techncal content and complex techncal means of fmgerprint feature extraction of wreless devices, under the premise of constant wireless space channel state, a method based on convolut i on neural network (CNN) for automat i c classification of wireless routers i s presented to solve the dffcult problem of fngerprnt extracton.