空间谱估计经典算法性能比较

空间谱估计是阵列信号处理的一个重要研究方向。空间谱估计理论与技术已日趋成熟，近几十年的经典谱估计技术包括：常规波束形成（CBF）、Capon谱估计、多重信号分类（MUSIC）、旋转不变子空间算法（ESPRIT）、最大似然（ML）、子空间拟合（SF），及这些算法的扩展和变形。上述算法在各个分散的文章中均有具体深入的理论分析和研究，亦有类似的2种或3种算法的性能比较，但是针对这些所有算法的性能比较，就笔者所知尚无公开报道，而使工程实现时对算法的选择没有依据。本文对这些经典算法做了简介，列出各个算法的优缺点，并对性能进行仿真比较，能直观地得到各个算法的性能对比，给工程实现算法选择提供理论依据。     

CCSDS近地LDPC译码算法研究

针对CCSDS（空间数据系统咨询委员会）推荐的近地LDPC（低密度奇偶校验）码技术进行了研究,建立了和积译码算法、对数似然和积译码算法、最小和译码算法的数学模型,并对上述译码算法的译码复杂度和译码性能进行了仿真分析.分析结果表明,和积译码算法与对数似然和积译码算法的译码性能距离香农限1.2 dB,最小和译码算法的译码性能距离香农限1.45 dB.因此,提出基于最小和译码算法的改进算法——偏移最小和译码算法与归一化最小和译码算法,并分析了这2种译码算法的译码复杂度,同时进行了大量仿真实验.实验结果表明,当偏移因子β=0.15时,偏移最小和译码算法性能达到最优,译码性能距离香农限1.25 dB;当归一化因子α=0.741 2时,归一化译码算法的译码性能达到最优,译码性能距离香农限1.2 dB.归一化译码算法具有优异的译码性能和合理的复杂度,可以遴选作为CCSDSLDPC的译码算法用于工程实现.此外,还研究了迭代次数对译码性能的影响,结果表明,当迭代次数大于10次时,译码性能提升不再明显,故工程实现时迭代次数应设置为10次.     

空间目标编目测量资源调度策略设计与实现

空间目标编目测量资源调度是一个复杂的系统问题。首先对测量资源调度要素进行分析,然后在对调度需求分解的基础上,给出一个基于CSP（约束满足问题）模型的随机搜索编目调度算法。该算法在目标数量多、任务数据量大、复杂度高的任务环境下,求解快速、稳定。最后对调度软件实现中的核心问题——数据结构组织方式进行了阐述,基于合理的数据结构,调度算法得以高效实现。     

宽带扩频阵列信号处理技术研究

提出了一种基于时域分析的复合加权宽带阵列信号处理新方法,该方法通过群时延、相位和幅度复合加权处理,可以避免复杂的频域聚焦算法,最后给出了处理方法的详细算法和实现框图。     

Implementation and convergence considerations of a linearlyconstrained adaptive array

O.L. Frost (1972) introduced a linearly constrained optimization algorithm that allows certain main beam properties to be preserved while good cancellation is attained. An open-loop implementation of this algorithm is developed. This implementation is shown to be equivalent to the technique developed by C.W. Jim (1977), L.J. Griffiths and C.W. Jin (1982), and K.M. Buckley and L.J. Griffiths (1982) whereby the constrained problem is reduced to an unconstrained problem. Analytical results are presented for the convergence rate when the sampled matrix inversion (SMI) or Gram-Schmidt (GS) algorithm are employed. It has been previously shown that the steady-state solution for the optimal weights is identical for both constrained and reduced unconstrained problems. It is shown that if the SMI or GS algorithm is employed, then the transient weighting vector solution for the constrained problem is identical to the equivalent transient weight vector solution for the reduced unconstrained implementation     

闭式叶盘叶身加工切触点规划算法研究与实现

在对闭式整体叶盘叶身造型方法分析的基础上,提出基于叶片防扭曲高质量造型的等误差切触点规划算法,详细描述了算法中边界参数点确定、切削行参数划分、走刀步长确定和型值点网格化四个步骤,给出了该算法在UGCAD系统中的实现及工程应用实例。     

基于属性形态学分析的图像显著极值检测及应用

由于图像噪声的存在,使得利用传统的极值检测算法通常会使要提取的显著极值淹没在大量的噪声极值中;同时由于先验知识的缺乏,采用普通滤波技术也往往不能很好的滤除噪声,反而会破坏图像的关键结构。本文提出了一种基于属性形态学分析的图像显著极值检测算法。该算法可以在不需要对图像进行滤波的前提下,从数学形态学的角度对图像极值的显著性进行计算和评估,从而能够较好地提取出显著的极值。在沉浸模拟算法的基础上,给出了基于属性形态学分析显著极值检测的快速算法实现,并将其成功应用在视觉注意选择和独立运动目标检测上。实践证明,该算法不仅具有较好的抗噪声特性,而且快速实用,具有广泛的应用价值。     

Interval Kalman filtering

The classical Kalman filtering technique is extended to interval linear systems with the same statistical assumptions on noise, for which the classical technique is no longer applicable. Necessary interval analysis, particularly the notion of interval expectation, is reviewed and introduced. The interval Kalman filter (IKF) is then derived, which has the same structure as the classical algorithm, using no additional analysis or computation from such as H/sup /spl infin//-mathematics. A suboptimal IKF is suggested next, for the purpose of real-time implementation. Finally, computer simulations are shown to compare the new interval Kalman filtering algorithm with the classical Kalman filtering scheme and some other existing robust Kalman filtering methods.     

一种DSMC方法的并行策略

为提高直接模拟蒙特卡罗（DSMC）仿真模拟的并行计算效率,基于消息传递接口（MPI）的并行环境,通过对比分析主从模式及对等模式两种程序设计模式下的并行效率,探讨了对等模式下非结构网格DSMC并行程序实现的关键技术及实施途径。提出了一种非结构网格下动态负载平衡DSMC仿真模拟的并行策略,设计了基于对等模式动态负载平衡的DSMC并行算法。最后以钝锥外形的高超声速绕流问题进行仿真模拟,验证本文并行算法的有效性,结果表明,本文设计的基于对等模式动态负载平衡的DSMC并行算法能够以高效的并行效率给出合理的结果。     

Adaptive detection threshold optimization for tracking in clutter

The adaptive optimization of detection thresholds for tracking in clutter is investigated for the probabilistic data association (PDA) filter. Earlier work on this problem by T.E. Fortmann et al. (1985) involved an approximate steady-state analysis of the state error covariance and is only suitable for time-invariant systems. Furthermore, the method requires numerous assumptions and approximations about the error covariance update equation, and uses a cumbersome graphical optimization algorithm. In this work we propose two adaptive schemes for threshold optimization, namely prior and posterior optimization algorithms which minimize the mean-square state estimation error over detection thresholds which depend on data up to the previous and current time-step, respectively. These algorithm are suitable for real-time implementation in time-varying systems. Some simulation results are presented     

Improving performance in pulse radar detection using neuralnetworks

A new approach using a multilayered feed forward neural network for pulse compression is presented. The 13 element Barker code was used as the signal code. In training this network, the extended Kalman filtering (EKF)-based learning algorithm which has faster convergence speed than the conventional backpropagation (BP) algorithm was used. This approach has yielded output peak signal to sidelobe ratios which are much superior to those obtained with the BP algorithm. Further, for use of this neural network for real time processing, parallel implementation of the EKF-based learning algorithm is indispensable. Therefore, parallel implementation has also been developed     

Decoding techniques in state estimation for dynamic systems withpast histories

States of dynamic models with a higher order memory are estimated using both a stack sequential decoding algorithm and the Viterbi decoding algorithm (VDA), without higher dimensional dynamic system representation. This results in memory reduction for state estimate implementation. It is found that state estimation with a stack sequential decoding algorithm is faster and more practical than the state estimation with the Viterbi decoding algorithm, even though the estimates obtained by the Viterbi decoding algorithm are superior     

基于基因算法的加工质量故障诊断研究与实现

以基于节约覆盖集理论的概率因果诊断模型为理论基础,以基因算法为该模型的求解策略,两者结合起来解决质量管理界的难点问题之一--机械加工过程质量故障诊断问题,其中又以构造一个合理有效的基因算法个体适合度评价函数为重点,同时引入了故障树分析方法,来提高诊断模型的准确性和可解释性。应用研究表明,该方法将有效地解决加工质量的诊断与改进问题。     

混合高斯测量噪声下的多模型滤波

就带有混合高斯测量噪声的离散时间系统 ,提出了一种简化的多模型滤波。理论分析证明该滤波方法用较少的计算量得到了与交互多模型滤波相同的估计性能。为满足应用要求 ,给出了该滤波器的数值鲁棒实现方法。一个关于仅有方位测量的制导例子验证了该滤波算法的有效性。     

导弹飞控系统可靠性仿真研究

提出了在几种主要因素影响下进行飞控系统性能和可靠性一体化分析的方法。建立了风、结构误差(包括弹体质量分布不对称、弹体不同轴、翼面安装角误差和舵面机械零位误差)、推力偏心和硬件故障等因素的扰动模型,给出了可靠性仿真的方法、实施步骤、具体算法流程和相关可靠性指标的计算公式。以某型导弹飞控系统作为案例,进行了大量仿真,分析了风等因素对该系统可靠性的影响。通过算例证明,所建立的扰动模型和可靠性分析方法是正确的。     

Aircraft Trajectory Control at the Motion on the Predetermined Route Based on the Global Navigation Satellite System

An algorithm to control the aircraft trajectory is proposed. This algorithm is based on the dynamic stochastic systems optimal control theory. The optimal control implementation is shown to reduce the deviation of the controlled trajectory from the predetermined one. The optimal control is based on estimating phase coordinates with the high accuracy by the global navigation satellite system.     

局部特征尺度分解改进算法

针对局部特征尺度分解估计包络曲线时未考虑其凹凸性的问题,提出了一种改进的算法。首先,利用3次样条插值方法进行均值点估计,结合前向以及后向插值2种方法,给出了双向形式;其次,参照标准算法,给出了具体的改进算法实现步骤;最后,通过仿真实例验证了有效性。     

用强耦合RANS方法模拟旋翼悬停流场

在旋转坐标系下,将Spalart-Allmaras(S-A)一方程湍流模型和Reynolds-averaged NavierStokes(RANS)方程耦合成一个新的RANS方程,并发展了基于多块重叠网格的强耦合RANS求解方法,用于直升机旋翼悬停流场的数值模拟.为了提高计算效率,针对多重网格方法在多块重叠网格上实施的困难,提出了一种基于重叠网格的多重网格实施方法.通过对Caradonna-Tung(C-T)和ONERA 7A旋翼悬停算例验证了发展的强耦合RANS方法和基于重叠网格的多重网格实施方法的有效性.研究结果表明:发展的基于重叠网格的多重网格方法有较高的计算效率,3层网格的加速比约为7.7;强耦合RANS法的计算精度明显高于传统的松耦合RANS方法,特别是在与阻力相关性能参数的预测中,强耦合RANS方法的预测结果更加精确.     

Sequential ridge regression

A sequential algorithm which closely approximates ridge regression is introduced, and it is pointed out that the desired sequential ridge estimator can be obtained by properly choosing the free parameters of a startup technique for ordinary sequential least squares estimation. The derivation of this result is trivial; however, since no matrix inversions are needed for its implementation, this algorithm is of practical importance     

A fast adaptive null-steering algorithm based on output powermeasurements

Presented and investigated here is a simple and fast adaptive algorithm for linear power inversion arrays whose nulls can be accurately steered by controlling the element weights. Based on measurements of the powers of the three signals derived from the array output and the output of an auxiliary beamformer, the algorithm tracks unknown jammers in the environment by steering the nulls of the array one by one in a cyclical time-multiplexed manner. When compared with the least mean square (LMS) algorithm, the proposed algorithm has about the same implementation complexity, a better convergence behavior and the advantage that nulls are directly available