MULTISTAGE ADAPTIVE HIGHER-ORDER NONLINEAR FINITE IMPULSE RESPONSE FILTERS FOR CHAOTIC TIME SERIES PREDICTIONS

A multistage adaptive higher-order nonlinear finite impulse response (MAHONFIR) filter is proposed to predict chaotic time series. Using this approach, we may readily derive the decoupled parallel algorithm for the adaptation of the coefficients of the MAHONFIR filter, to guarantee a more rapid convergence of the adaptive weights to their optimal values. Numerical simulation results show that the MAHONFIR filters proposed here illustrate a very good performance for making an adaptive prediction of chaotic time series.     

基于回溯筛选的稀疏重构时延估计算法

针对无线定位中时延估计在小样本(单快拍)、低信噪比条件下需要大量独立分布测量数据问题,提出了一种基于回溯筛选的稀疏重构时延估计算法,实现了单快拍、低信噪比条件下接收信号的精确时延估计.该算法首先建立接收信号的稀疏表示模型,然后基于该模型建立正交观测矩阵,最后在重构算法中引入回溯筛选思想,利用时延与观测矩阵之间的一一对应关系得到时延的无偏估计.对该模型下时延估计的克拉美罗界进行了推导.仿真分析表明,所提方法在单快拍、低信噪比条件下精度远高于求根多重信号分类算法,相比于正交匹配追踪算法,在较小的复杂度代价下性能得到了较大提升.     

互谱域MVDR自适应时延估计方法研究

针对窄带信号,通过构造互谱时间序列,在互谱域建立了平稳时间序列时延估计的最小方差无畸变响应(MVDR)滤波器模型;利用分段近似处理,类比空间MVDR自适应算法,给出了其具体算法(Algorithm of MVDR in cross spectral domain,CSMVDR);进行了数值仿真实验研究和海上实验数据处理。数值仿真与实验数据处理结果初步验证了CSMVDR时延估计对于舰船辐射噪声的适用性,CSMVDR时延估计有比相关检测更好的时延估计性能,能够提高信噪比增益和时延估计精度。     

一种基于最优频段的X射线脉冲星累积轮廓时延估计方法

为提高X射线脉冲星导航中累积脉冲轮廓的时间延迟估计精度, 分析了X射线脉冲星累积脉冲轮廓的频谱特性和现有Taylor快速傅立叶变换时延估计算法的缺陷, 提出了一种基于最优频段的累积轮廓时延估计算法, 并通过建立不同信噪比下时延估计误差与所采用频段之间的关系以确定最优频段. 数值及实测数据实验结果表明: 在短时观测或光子流量较小时, 该算法优于常用的近似最大似然 (FAML)、相关 (CC)、最小二乘 (NLS) 及加权最小二乘 (WNLS) 方法; 在观测时间较长或光子流量较大时, 该算法的估计精度与CC及NLS方法相当, 但其运算量低于NLS, FAML 及WNLS方法. 本文所提算法适用于短时观测脉冲轮廓或低流量脉冲星的高精度时延估计.     

多频带期望值最大声信号时延估计EI北大核心CSCD

通过分析复杂环境中不同频带声信号时延估计的差异,提出多频带期望值最大时延估计方法。为了使各频带之间无重叠,该方法采用独立分带划分声信号不同频带,然后计算各频带广义互相关函数,并对子带广义互相关函数建立最大似然模型,最后利用期望值最大算法将多维优化转为一维优化的迭代式,获得最优子带广义互相关函数,在此基础上估计声信号的时延信息。数据仿真和实际实验结果表明,多频带期望值最大化时延估计相较常规时延估计有效估计值的百分比提升了10%,并将最优频带互相关函数应用到该定位算法中,在网格间距为0.3 m时,得到的峰值区域汇聚更明显,定位效果更好。     

具有不同时延的多智能体系统一致性分析

针对有向加权且存在全局可达点的静态网络拓扑, 考虑了同时具有通信时延和输入时延的一阶、二阶智能体系统的运动一致性问题. 基于广义Nyquist准则与频域控制理论的方法, 分析并得到了网络中所有智能体渐进收敛到一致状态的充分条件. 通过该条件发现一致性的达到只与系统的耦合强度、智能体的输入时延以及各自的连接状态信息有关, 与通信时延无关. 但是, 通信时延的存在却要影响系统的动态特性. 仿真实验结果进一步验证了理论分析所得结论的正确性.     

基于滚动时域估计的飞行器姿态估计及三轴磁强计在线校正

针对飞行器姿态估计以及三轴磁强计在线校正问题, 提出了一种实时滚动时域估计算法. 首先, 为了解决在卡尔曼滤波框架下系统约束不能显式求解的问题, 设计了滚动时域估计滤波算法. 该算法将飞行器姿态估计问题转化为优化问题, 显式求解四元数归一化性质, 缩小搜索空间的同时提高了搜索效率和精度. 其次, 滤波时域窗内应用高斯-牛顿迭代法求解最优状态估计值, 满足了实时性要求. 最后, 在没有增加系统状态维数的情况下, 在线求解了三轴磁强计校正参数, 保证了磁强计量测值以矢量形式输入系统. 仿真结果表明, 由于合理地利用了历史信息, 该方法精度较高, 且对初始误差、系统误差均不敏感, 具有一定鲁棒性.     

基于改进MCA的干涉高光谱图像分解

干涉高光谱图像特殊的成像原理,使其帧内存在着大幅值且位置固定的干涉条纹,而帧间存在着水平移位的背景图像,这种特点会严重的破坏原始图像的固有结构,从而导致新兴的压缩感知理论与传统压缩算法的直接应用无法得到理想的效果。由于干涉条纹信息与背景图像信息的特征不同,能够对干涉条纹与背景图像进行稀疏表示的正交基也是不同的。基于这种思想,使用MCA(morphological component analysis)算法对干涉高光谱图像中干涉条纹信息与背景图像信息进行分离处理。由于干涉高光谱图像数据量庞大,传统的MCA算法对干涉高光谱数据的图像分解,迭代收敛速度慢,运算效率较低,故而针对干涉高光谱数据特点对传统MCA算法进行改进,改变其迭代收敛条件,当分离后的图像信号与原始图像信号的误差已经基本保持不变时,即终止迭代;并根据对应正交基能且仅能稀疏表示对应信号的思想,对阈值采用自适应的方式进行更新,在新的阈值更新模式中,图像信号在不同正交基下的映射系数被计算与比较。大量实验结果表明,对于LASIS数据与LAMIS数据,MCA算法都能够较完美的将干涉高光谱图像分解,改进的MCA算法更能在保持完美分解输出结果的同时,相对于传统MCA方法显著的减小迭代次数,更快的达到迭代收敛条件,从而有效的提高了算法的运算效率与实时性需求,也为新兴的压缩感知理论在干涉高光谱图像中的进一步应用提供了一种很好的解决方案。     

一种基于三阶Volterra滤波器的混沌时间序列自适应预测方法

基于Takens的相空间延迟坐标重构,研究了用于混沌信号预测的三阶Volterra滤波器的一种乘积耦合近似实现结构,并应用于典型的低维混沌时间序列和具有高维混沌特性的EEG信号的预测.数值研究表明:这种滤波器结构对于低维混沌时间序列的预测精度可以比二阶Volterra滤波器提高103倍,而且能够较好地对一些具有高维混沌特性的EEG信号进行预测 关键词： 混沌 非线性自适应预测 三阶Volterra滤波器 electroencephalography信号     

快速收敛最小方差无畸变响应算法研究及应用

常规最小方差无畸变响应(MVDR)自适应波束形成是一种高分辨窄带波束形成器,它是利用实际声场的窄带互谱密度矩阵(CSDM)估计出自适应波束形成权向量。在实际应用中,MVDR算法需要较长的观测时间估计协方差矩阵,不利于对高速运动目标进行定位;对于宽带目标信号,MVDR算法需要对每一个CSDM进行求逆运算,计算量较大;在相干源条件下,目标信号之间会发生"对消"现象,MVDR算法性能急剧恶化。本文提出了基于子带子阵处理的快速收敛MVDR自适应波束形成方法。首先将全频带划分成一组子带,将接收线阵划分成一组子阵,然后对每一子带计算降维的驾驶协方差矩阵(STCM),从而得到快速收敛MVDR自适应波束形成的权值和空间谱估计结果。同时采用双向空间平滑方法对相干源进行MVDR空间谱估计。仿真和海试数据处理结果表明该算法在保证高分辨力的同时,具有瞬时收敛的性能,双向空间平滑技术具有良好的解相干性能。      

一种新的卫星钟差Kalman滤波噪声协方差估计方法

采用Kalman滤波方法进行钟差参数计算和预报时, 需确定Kalman滤波噪声协方差矩阵. 针对这一问题, 提出了一种新的卫星钟差Kalman滤波噪声协方差估计方法, 通过建立新息的相关函数序列与未知的噪声参数间的线性函数模型, 采用最小二乘法进行噪声参数估计. 采用精密钟差数据进行钟差参数估计和预报分析, 结果表明, 该方法具有较好的收敛性, 并与顾及随机噪声模型的开窗分类因子自适应抗差估计方法进行对比分析, 验证了新方法的正确性和有效性.     

Policy iteration optimal tracking control for chaotic systems by using an adaptive dynamic programming approach

A policy iteration algorithm of adaptive dynamic programming(ADP) is developed to solve the optimal tracking control for a class of discrete-time chaotic systems. By system transformations, the optimal tracking problem is transformed into an optimal regulation one. The policy iteration algorithm for discrete-time chaotic systems is first described. Then,the convergence and admissibility properties of the developed policy iteration algorithm are presented, which show that the transformed chaotic system can be stabilized under an arbitrary iterative control law and the iterative performance index function simultaneously converges to the optimum. By implementing the policy iteration algorithm via neural networks,the developed optimal tracking control scheme for chaotic systems is verified by a simulation.     

Passive tracking of a moving source by a single observer in shallow water

Tracking of a moving source (MS) in shallow water is analyzed. The noisy observations of bearing, β, and any two time delays, τ₁ and τ₂, are made by a single observer, as opposed to spatially separated observers. Here, the time delays, τ₁ and τ₂, are, respectively, the differences in arrival times between the surface reflected path and direct path signals, and between the bottom reflected path and direct path signals [1]. The estimation process of MS kinematics is greatly aided by the two time delay measurements. These measurements impart consistency to the observation equations at each instant of time. In contrast to noisy bearings-only motion analysis, tracking is performed in a three-dimensional frame with concomitant advantages. Principally, the unwieldy requirement for a velocity change, anterior to convergence of the Kalman filter, is eliminated. The convergence time for the filter is greatly diminished. Furthermore, the filter displays a low operational threshold. Its stability is maintained when mismodeling exists, either in the observation or in the kinematical models.     

混沌时间序列的自适应高阶非线性滤波预测

根据混沌序列产生的确定性和非线性机制,基于Volterra级数展式和混沌序列高阶奇异谱特征,提出了一种高阶非线性傅里叶红外(HONFIR)滤波预测模型用于混沌时间序列的自适应预测.其自适应算法采用时域正交算法来自适应地跟踪混沌的运动轨迹,而不是重构混沌系统 的全局或局部运动轨迹.实验研究表明：(1)这种HONFIR自适应滤波器能够有效地预测一些超 混沌序列.(2)预测混沌序列的性能与预测模型的非线性拟合能力有关,但并非非线性程度越 高,预测性能就越好.(3)当HONFIR滤波器对混沌序列的非线性拟合精度高时,其自适应预测 的性能与其输入维数的关系不受Takens嵌入定理的约束.(4)HONFIR自适应滤波器具有一定的 抗噪能力. 关键词： 混沌 非线性模型 滤波器     

Performance evaluation of eigendecomposition-based adaptive clutter filter for color flow imaging

In this paper, a newly developed eigendecomposition-based adaptive filter is applied in flow velocity estimation and its performance is evaluated and compared with the static polynomial regression filter. Results demonstrate that the eigendecomposition-based filter is the fully-adaptive and can not only suppress the clutter signals with minimal loss of low-flow signal, but also give less influence on the flow velocity estimation especially under the circumstance of high clutter velocity.     

基于二次相关的语音信号时延估计改进算法

目前语音信号的时延估计研究,大部分采用的是广义互相关算法。然而,广义互相关时延估计算法易受噪声和混响环境影响。为此,本文提出了一种基于二次相关的语音信号时延估计改进算法,该算法对语音信号进行二次互相关运算,并结合Hilbert变换,对二次互相关峰值进行进一步的锐化处理,使得反映时延的峰值点检测更为准确。实验结果表明,改进的时延估计方法在非平稳的语音信号中能够有效地抑制噪声干扰,且在不同混响条件下时延估计具有更好的性能。     

一种改进的非整数自适应时延估计方法*

针对基于时延估计的机电设备故障声定位中的低信噪比和脉冲噪声情况,用α稳定分布建模噪声,改进了非整数自适应时延估计方法。共变相关法对观测序列进行时延估计粗测,将得到的估计值作为非整数自适应时延估计器的初值;将共变相关法中共变序列作为时延估计器的输入信号,在最小平均p范数准则下迭代得到非整数时延估计值。共变序列保留了原始序列间的时延信息,削弱了不相关的噪声。计算机仿真对比实验验证了改进的方法在脉冲环境和低信噪比条件下有更好的性能。     

A robust adaptive hybrid feedback cancellation scheme for hearing aids in the presence of outliers

It has been demonstrated that the Filtered-x Wilcoxon LMS (FxWLMS) based adaptive filter mitigates the effect of the outliers acquired by the microphone signal of hearing aids by minimizing the Wilcoxon norm and hence shows better cancellation performance than the existing Filtered-x LMS (FxLMS) algorithm. The prediction error method based adaptive feedback canceller (PEMAFC) reduces the bias present in the estimate of the feedback path due to the continuous adaptive filtering (CAF). However, the impulse response of the measured feedback path is close to zero for the first many samples due to the delay introduced by ADC converters and then contains few significant values, which results in slow convergence rate when an adaptive filter is used to model the same. To overcome this limitation, we propose a proportionate normalized WLMS (PNWLMS) algorithm based PEMAFC (P-PNWLMS) for feedback cancellation in hearing aid in the presence of outliers. Further, with an objective to improve the convergence rate and performance accuracy simultaneously, this paper proposes a novel convex PNWLMS (CPNWLMS) algorithm which incorporates convex combination of PNWLMS and WLMS algorithms. The weight update equations are derived for PEMAFC trained by PNWLMS (P-PNWLMS) and CPNWLMS (P-CPNWLMS) algorithms respectively. The results of the simulation study show improved performance of the proposed CPNWLMS based adaptive filter over its component filters.     

Real time inverse filter focusing through iterative time reversal

In order to achieve an optimal focusing through heterogeneous media we need to build the inverse filter of the propagation operator. Time reversal is an easy and robust way to achieve such an inverse filter in nondissipative media. However, as soon as losses appear in the medium, time reversal is not equivalent to the inverse filter anymore. Consequently, it does not produce the optimal focusing and beam degradations may appear. In such cases, we showed in previous works that the optimal focusing can be recovered by using the so-called spatiotemporal inverse filter technique. This process requires the presence of a complete set of receivers inside the medium. It allows one to reach the optimal focusing even in extreme situations such as ultrasonic focusing through human skull or audible sound focusing in strongly reverberant rooms. But, this technique is time consuming and implied fastidious numerical calculations. In this paper we propose a new way to process this inverse filter focusing technique in real time and without any calculation. The new process is based on iterative time reversal process. Contrary to the classical inverse filter technique, this iteration does not require any computation and achieves the inverse filter in an experimental way using wave propagation instead of computational power. The convergence from time reversal to inverse filter during the iterative process is theoretically explained. Finally, the feasibility of this iterative technique is experimentally demonstrated for ultrasound applications.     

基于粒子滤波的单通道正弦调频混合信号分离与参数估计

以单通道正弦调频(SFM)混合信号为研究对象,提出了基于粒子滤波的正弦调频混合信号分离与参数提取方法.针对正弦调频混合信号频率无跳变的特征,提出了一种基于粒子滤波的相位差解混叠算法,并通过源信号相位差解决了本算法中粒子滤波高维状态空间降维问题,提出了一种适合高维状态空间的似然函数模型,比较固定长度粒子估计值和真实值误差,进而准确衡量粒子权重.通过在重采样后引入MCMC转移,解决了静止参数下粒子多样性降低问题,有效提高粒子滤波迭代收敛速度.从而在先验知识仅已知信号调制方式的情况下,完成对单通道正弦调频混合信号的参数提取,并通过重构信号完成正弦调频混合信号分离.最后通过仿真分析发现,该方法能够有效的实现正弦调频混合信号的分离与参数估计.