ID 原文 译文
6024 最后通过改进K-Means算法完成聚类分选。 finally by improving the K - Means complete clustering sorting algorithm.
6025 仿真实验表明:所提方法能够应对存在频率捷变,重频参差、抖动、参数交叠、局部脉冲丢失的复杂信号环境,分选正确率明显提升。 Simulation experiments show that the proposed method can deal with frequency agility, heavy frequency varies, jitter, parameters, overlapping, local pulse missing complex signal environment, sorting accuracy improved significantly.
6026 武器装备性能评估贯穿于装备发展的全寿命过程,对于装备定型与实战化运用具有重要意义。 Weapons and equipment performance evaluation throughout the whole life of equipment development process, is of great significance for the equipment to finalize the design and application of the combat.
6027 针对传统反向传播(back propagation,BP)神经网络模型易陷入局部最优、武器装备评估数据少等问题,提出了基于数据包络分析和Bootstrap法的改进型BP神经网络模型。 For traditional back propagation (BP) back propagation, a neural network model is easy to fall into local optimum, weapons and equipment assessment data problems such as less, was proposed based on data envelopment analysis (dea) and modified BP neural network model of the Bootstrap method.
6028 利用数据包络分析处理原始指标得到3项优化参数,结合Bootstrap法对其进行扩充,再通过BP神经网络建立评估模型。 By using data envelopment analysis (dea) to deal with the original indexes have three optimization parameters, combined with the Bootstrap method to expand, again by the BP neural network assessment model is established.
6029 实验表明,改进模型得到的合成孔径雷达(systemic aperture radar,SAR)导引头测高性能评估结果,其决定系数和误差系数均有明显改善。 Experiments show that the improved model of synthetic aperture radar (systemic aperture radar, SAR) seeker performance evaluation results, the decision coefficient and error coefficient were obviously improved.
6030 该模型不仅规避了模糊综合评判法主观性强、精确度不高等问题,同时有效解决了传统BP神经网络模型易陷入局部最优和武器装备评估数据少的两个问题。 The model not only avoid the fuzzy comprehensive evaluation method of subjectivity is strong, the accuracy is not high, at the same time effectively solves the traditional BP neural network model into the local optimum and weapon equipment evaluation data less two problems.
6031 为解决起伏地形环境下异构多传感器网络表面覆盖问题,提出了多传感器多阶段分簇部署方法。 To solve the heterogeneous sensor network covered more undulating terrain environment problems, puts forward the multisensor multi-stage cluster deployment method.
6032 首先,考虑地形遮挡效应,给出了起伏地形环境下的传感器侦察与通信模型。 First, consider the terrain shade effect, gives the undulating terrain environment sensor reconnaissance and communication model.
6033 其次,为快速获取传感器覆盖范围,构建了基于视线交点相似性判断的改进型d-Xdraw可视域求解算法。 Second, to quickly get the sensor coverage, build similarity judgment based on the view point of the modified d - Xdraw horizon algorithm.