ID 原文 译文
57408 基于参数联合估计的假目标鉴别方法可通过增加雷达站数量来提高假目标鉴别概率，然而，过度增加雷达站数量会造成设备资源的严重浪费. For the false target identification method based on joint estimation of parameters，the discrimi- nation of false target probability can be improved by increasing the number of radar stations. However， the excessive increased radar stations will cause a serious waste of equipment resources.
57409 对此，提出基于多站雷达系统假目标鉴别过程中渐进收缩的子站选择策略. For this prob- lem，a gradual shrinkage subset selection strategy on multiple-radar anti-jamming systems is proposed.
57410 对于空间已有的雷达站，在满足预设假目标鉴别性能的前提下，考虑通过快速收缩和全局收缩 2 种筛选方式，迭代选出系统中空间分布更有优势、鉴别能力更强的发射或接收站，共同组成雷达子站. Aiming to existing radar stations，the rapid shrinkage method and the global shrinkage method are consid- ered to select some transmitting and receiving stations to form the radar subset which guarantee the preset false target discrimination performance. All of the selecting stations have better spatial distribution or stronger discrimination ability in the system.
57411 相比于穷举搜索方法，子站选择策略可大幅降低筛选过程的时间复杂度. Compared with exhaustive search，the proposed subset se- lection strategy has a great reduction in computational complexity.
57412 仿真结果表明，子站能够保持与原多站雷达系统近似的鉴别效果，同时优化了雷达设备数量，减少了融合中心处理的数据量和所需的通信链路，有效节约了运作成本 Simulation shows that the radar subset can maintain the similar discrimination performance with the original multiple-radar systems. At the same time，it optimizes the number of radar stations，reduces the amount of data processed by the fusion center and the required communication links，which effectively save the operating cost.
57413 针对交通流量特性和外部因素对交通流量预测结果的影响，提出了一种对城市短时交通流量预测的模型CNN-ＲesNet-LSTM，将卷积神经网络( CNN) 、残差神经单元( ＲesNet) 和长短期记忆循环神经网络( LSTM) 集成到一个端到端的网络框架. With the continuous advancement of smart city construction，urban short-term traffic flow fore?casting becomes more and more important. According to the influence of traffic flow characteristics andexternal factors on traffic flow forecast results，the model CNN-ＲesNet-LSTM for urban short-term trafficflow forecasting is proposed. The model integrates convolutional neural networks( CNN) ，residual neuralunits( ＲesNet) and long-short-term memory networks( LSTM) into an end-to-end network framework.
57414 利用卷积神经网络来捕获城市区域间交通流量的局部空间特征，并在卷积神经网络中加入多个残差神经单元来加深网络深度，可提高预测的准确性; The convolutional neural network is used to capture the local spatial characteristics of traffic flow，andmultiple residual neural units are added to deepen the network depth and improve the prediction accura?cy.
57415 利用长短期记忆循环神经网络来捕获交通流量数据的时间特征; On the other hand，the long short-term memory-cycle neural network is used to capture temporalcharacteristics of traffic flow data.
57416 利用相应的权重将 2 个网络的输出结果融合，得到通过轨迹数据预测的结果; The output results of the two networks are combined by the correspond?ing weights to obtain the predicted results through the trajectory data.
57417 最后与外部因素融合，得到城市区域的交通流量预测值. Finally，the traffic flow predictionvalues of the urban areas are obtained by fusing with external factors.