ID	原文	译文
7704	针对相控阵雷达导引头由于前向通道增益和波束控制增益刻度尺度不同引起的扩展卡尔曼滤波(extended Kalman filter,EKF)去耦中误差量过大的问题,提出了基于粒子群优化的EKF去耦算法。	For phased array radar seeker because the forward channel gain and beam caused by different control gain calibration scale extended Kalman filtering (extended Kalman filter and EKF) the problem of the large amount of error in the decoupling, EKF decoupling based on particle swarm optimization algorithm is proposed.
7705	采用了最小均方差为适应度函数,对两个增益参数进行组合优化。	Adopt the minimum mean square error as the fitness function, to combinatorial optimization of two gain parameters.
7706	然后通过建立EKF的系统模型,推导了提取的视线角速率与增益参数之间的关系,使得滤波后的估计值为最优的后验估计。	Then through the establishment of system model of EKF, extraction of line of sight angular rate and the gain parameters was deduced, the relationship between the make the estimate after filtering for optimal a posteriori estimates.
7707	最后,通过仿真表明该算法可以很好地解决误差量过大的问题,并验证了所提算法在相控阵雷达导引头去耦和视线角速率提取中的有效性。	Finally, the simulation shows that the algorithm is a good way to solve the problem of the large amount of error, and verify that the proposed algorithm in phased array radar seeker decoupling and effectiveness in the line of sight angular rate extraction.
7708	目标在定位空间中具有稀疏特性,基于该特点提出了一种稀疏重构的时延定位算法;	Target in the location space with sparse features, based on the characteristics of the proposed a sparse reconstruction time delay of the localization algorithm;
7709	已有的来波到达时间(time-of-arrival,TOA)算法大部分只利用了单次TOA进行估计,其定位结果受噪声影响较大,因此进一步提出对多样本的到达时间进行联合估计,从而提高算法对噪声的稳健性,并提高算法的定位精度。	Existing wave arrival time (the time - of - concatenated, TOA) most only using a single TOA estimation algorithm, the locating results are greatly influenced by noise, thus further put forward to the arrival time for joint estimation of diverse, thus improve the algorithm robustness to noise, and improve the positioning accuracy of the algorithm.
7710	与已有算法相比,所提算法的优点是定位精度更高,对噪声有更强的稳健性。	Compared with the existing algorithms, the proposed algorithm has the advantage of location accuracy is higher, have stronger robustness for noise.
7711	仿真结果验证了所提算法的有效性。	The simulation results verify the effectiveness of the proposed algorithm.
7712	传统基于离散时间贝叶斯网络的动态故障树分析方法的计算时间和计算精度受时间分段影响极大。	Traditional discrete time dynamic bayesian network based fault tree analysis method of computing time and calculation accuracy affected by time segmentation greatly.
7713	基于复合梯形积分方法分析传统方法的计算误差,提出改良的动态门转化方法,补偿其计算误差。	Based on the complex trapezoidal integral method to analyze the calculation error of the traditional method, put forward improved dynamic door transformation method, the calculation error compensation.