ID 原文 译文
6134 最后,对目标空间谱的变化过程采用隐马尔可夫模型进行描述,并将空间谱连续慢变的客观规律应用到目标信号超参数的概率密度计算中,构建基于多个时刻阵列信号模型的空间谱稀疏重构模型。 Finally, the change process of space target spectrum using hidden markov model is described, and the application of the objective laws of spatial spectrum slowly varying continuously to the target signal super parameters in calculating the probability density of spatial spectrum of array signal model based on multiple time sparse reconstruction model.
6135 计算机仿真研究和海试数据验证结果表明:所提方法在拖曳线列阵机动条件下,能够有效抑制固有的左右舷模糊,同时具有更好的重构精度,从而实现拖曳线列阵空间谱的优效估计。 Computer simulation and the sea trial data validation results show that the proposed method under the condition of the towed line array motor, can effectively restrain the port/starboard blur, inherent has better reconstruction accuracy at the same time, so as to realize the optimal effect of towed line array spatial spectral estimation.
6136 为降低传统仿真优化方法所需的仿真次数,从而缩短仿真优化时间,提出了基于广义回归神经网络(generalized regression neural network,GRNN)的仿真优化算法设计。 To reduce the number of traditional simulation optimization method for simulation, thus shortening the time of simulation optimization, was proposed based on generalized regression neural network (generalized regression neural network, the GRNN) simulation optimization algorithm design.
6137 首先,利用仿真生成一定数量的样本集,利用GRNN进行训练,得到初始回归曲面,并在该曲面上利用模式搜索算法找出全部可能的局部极小,由于可能会找到一些假局部极小点——噪声点,设计了剔除噪声点的方法,得到全部局部极小; First of all, the use of simulation to generate a certain amount of sample set, using GRNN training, get the initial return to surface, and on the surface by using pattern search algorithm to find all the possible local minimum, because may find some false local minimum points, noise, designed to eliminate the noise points method, got all the local minimum;
6138 在各局部极小点周围增补少量仿真样本,再次利用GRNN进行训练,得到新的回归曲面。 In each local minimum point around supplemented small samples of simulation, using GRNN training again, get new regression surface.
6139 重复增补样本,直到得到仿真优化的最优解。 Repeat supplemented sample until get the optimal solution of simulation optimization.
6140 实例表明,所提方法能够有效降低所需样本的数量,实现仿真优化问题的求解。 Examples show that the proposed method can effectively reduce the number of samples required, to achieve the simulation solution of the optimization problem.
6141 针对滤波器组多载波(filter bank multi-carrier,FBMC)信号的符号周期的盲估计问题,提出了自相关二阶矩和循环自相关算法。 For filter set of multicarrier (filter bank multi - carrier, FBMC) signal cycle of symbols of blind estimation problem, put forward the autocorrelation second moment and cyclic autocorrelation algorithm.
6142 首先通过滤波器组多载波的交错正交幅度调制(FBMC-offset quadrature amplitude modulation,FBMC-OQAM)信号的自相关二阶矩特性,估计出该信号的符号周期; First through the filter of multicarrier interlacing quadrature amplitude modulation (FBMC - offset quadrature amplitude modulation, FBMC - OQAM) signal autocorrelation second moment characteristics, estimate the signal of the symbol cycle;
6143 然后根据FBMC-OQAM信号的循环自相关在循环频率α=0切面具有离散谱线特征进行符号周期估计; Then according to the FBMC - OQAM signal of cyclic autocorrelation in cyclic frequency = 0 section has the characteristics of discrete spectrum line is used to estimate symbol cycle;