ID 原文 译文
55557 在惯性网络结构分析基础上,建立了惯性网络动态测量模型,采用最大似然估计和信息滤波法,设计了分布式惯性测量融合与导航状态融合的分阶段信息融合算法,通过仿真进行了验证。 In inertial network structure analysis, based on the inertial network dynamic measurement model is established, using the maximum likelihood estimation and information filtering method, and designs the distributed fusion inertial measurement and navigation fusion in stages of information fusion algorithm, and is verified by simulation.
55558 结果表明,所提方法充分利用了其他节点的惯性传感器信息和同类导航状态信息,可以提高整个惯性传感器网络的估计性能和故障容错能力,在提高低性能节点导航精度的同时实现对高性能节点的自动对准。 The results show that the proposed method can make full use of the other nodes of the inertial sensor information and similar navigational state information, can improve the estimation performance of the inertial sensor networks and fault tolerance, while improve navigation accuracy is low performance for high performance automatic alignment of nodes.
55559 微多普勒特征是运动目标的独特特征,可为目标分类识别提供依据。 Micro-doppler features are the unique characteristics of moving targets, can provide the basis for target classification.
55560 人体行走是典型的非刚体运动,其雷达回波中包含多个频率分量,且分量之间频率重叠严重,因此对其进行微多普勒特征分析难度较大。 Human walking is a typical non rigid motion, the radar echo include multiple frequency components, and components between frequency overlap is serious, so the analysis of micro-doppler features is difficult.
55561 首先构建人体行走模型及其雷达回波模型,然后以广义S变换为分析工具,对人体主要部位、单人行走和两人行走进行微多普勒特征分析,提出应该以脚部、小腿和躯干部的微多普勒特征作为人体行走识别的主要特征,为后续深入研究人体微多普勒特征奠定了基础。 Human walking model was constructed and its radar echo model first, and then generalized S transform as the analysis tool, walk to the main points, single and two people walk micro-doppler features analysis, proposed should to the feet, legs and body cadres of micro-doppler features as the main characteristics of the human walking recognition, for further in-depth study of human body micro-doppler features laid a solid foundation.
55562 大规模多输入多输出(multiple input multiple output,MIMO)中基站天线数远大于用户数,固匹配滤波(matching-filtering,MF)检测子在上行多用户检测中有重要的应用潜力。 Large-scale multiple input multiple output (multiple input multiple output, MIMO) in the base station antenna number greater than the number of users, solid matched filtering (matching filtering, MF) child in the uplink multiuser detection has important application potential.
55563 但是和传统的迫零(zero-forcing,ZF)、最小均方误差(minimum mean square error,MMSE)检测子相比,MF存在明显的性能损失,且在高信噪比区域出现"误差门限"效应。 But forced and traditional zero (zero - forcing, ZF), minimum mean square error (minimum mean square error, MMSE) detection compared, MF has significant performance losses, and in a high SNR region "error threshold effect".
55564 为了解决MF上述的不足,提出一种基于匹配滤波连续干扰相消和局部搜索的多用户检测算法。 In order to solve the MF these deficiencies, the paper puts forward a kind of successive interference cancellation based on matching filtering and multi-user detection algorithm of local search.
55565 理论分析表明,与MF类似,所提方法的计算复杂度远低于MMSE、ZF,并且实验证明此方法的误符号率和误比特率明显低于MF。 Theoretical analysis shows that similar to MF, proposed method of computational complexity is much lower than MMSE, government, and the experiment prove that this method of symbol error rate and bit error rate is significantly lower than the MF.
55566 现有计算渐消因子的自适应卡尔曼滤波器得到的通常是标量渐消因子,从而导致各滤波通道具有相同的调节能力,不利于提高滤波精度。 Existing calculation fading factor of adaptive kalman filter is usually get a scalar fading factor, which can lead to the regulation of the filtering channel with the same capacity, is not conducive to improve the filtering accuracy.