ID 原文 译文
50807 证明结果表明，由服从某种特定概率分布的项构造的Toeplitz矩阵以较大概率满足有限等距特性。 prove the results showed that the by obeying a certain probability distribution of term structure of Toeplitz matrices with limited isometric characteristics for large probability.
50808 最后，对最小二乘算法(least square，LS)、线性最小均方误差(linear minimum mean square error，LMMSE)算法和高斯观测矩阵的压缩感知算法以及Toeplitz观测矩阵的压缩感知算法进行了对比分析，Toeplitz观测矩阵的压缩感知算法在性能方面要优于高斯观测矩阵的压缩感知算法和传统算法， Finally, the least squares algorithm (further square, LS), linear minimum mean square error (linear minimum mean square error, LMMSE) algorithm and gaussian observation matrix compression perception and observation of Toeplitz matrices of compressed sensing algorithm through the contrast analysis of observation of Toeplitz matrices of compressed sensing algorithm in terms of performance is better than gaussian observation matrix of compressed sensing algorithm and traditional algorithm,
50809 运算复杂度方面要优于高斯随机矩阵， the computation complexity is better than gaussian random matrix,
50810 为压缩感知实现无失真地重构原始信号提供了理论和应用参考。 to implement undistorted compression perception to reconstruct the original signal provides the theoretical reference and application.
50811 针对现有多个弱小目标检测前跟踪(track-before-detect，TBD)算法存在的跟踪精度低，算法复杂度高等问题，提出一种新的基于概率假设密度(probability hypothesis density，PHD)的TBD算法。 In view of the existing multiple weak target detection before tracking (track - before - detect, TBD) algorithm is low tracking precision, the algorithm complexity is high, put forward a new hypothesis is based on probability density (aim-listed probability content, density, PHD) TBD algorithm.
50812 所提算法通过高斯粒子滤波对PHD中的各高斯项进行递归运算、进行多帧能量累积， Proposed algorithm by gaussian particle filter is applied to the each gaussian PHD to recursive computation, to frame more energy accumulation,
50813 并提取高斯项的均值为目标的状态，达到检测与跟踪多个弱小目标的目的。 and extract the gaussian item as the target state, to achieve the purpose of detecting and tracking multiple weak small targets.
50814 算法在随机集滤波框架下完成未知数目的多个弱小目标跟踪，不仅充分利用粒子滤波的非线性估计能力，同时避免了传统算法利用模糊聚类进行目标状态提取所带来的跟踪精度低等问题。 Algorithm in random set filter framework to complete multiple weak target tracking unknown purpose, not only make full use of the nonlinear estimation of particle filter ability, at the same time to avoid the traditional algorithm using fuzzy clustering to extract target state caused by the problem of low accuracy of tracking.
50815 仿真结果表明，所提算法与传统方法相比，在降低算法复杂度的同时，对多个红外弱小目标具有更加良好的实时检测和跟踪性能。 The simulation results show that the proposed algorithm is compared with the traditional method, at the same time, reduce the complexity of the algorithm for multiple infrared weak small targets with more good real-time detection and tracking performance.
50816 针对复杂环境下自主水下航行器(autonomous underwater vehicle，AUV)组合导航系统中存在噪声不确定或者易发生变化的情况， In view of the complex environment of autonomous underwater vehicle (autonomous underwater vehicle, the AUV) integrated navigation system exist in the noise uncertainty or easy to change,