基于格型预处理的二阶自适应Volterra滤波器

虽然Volterra LMS(VLMS)算法简单,计算量小,但是由于Volterra滤波器系统是非线性的,使得输入向量自相关矩阵包含了输入信号的高阶统计量,导致矩阵特征值扩展很大,因此其收敛速度一般很慢.为加快收敛速度,利用格型滤波器后向预测误差的正交特性,提出采用两个格型滤波器正交预处理的自适应Volterra滤波算法.仿真结果表明,该算法具有比VLMS算法和GPV算法更为优越的收敛性能.     

一种新的带窗重叠自适应滤波器

基于一种带窗重叠自适应滤波器，将重叠滤波思想引入LMS算法。利用重叠滤波的平滑性，将加窗重叠滤波和LMS算法相结合，给出了窗加权重叠LMS(WO-LMS)算法。与传统的LMS算法相比，WO-LMS算法既提高了收敛速度又可以得到较低的稳态均方误差。理论分析了算法的收敛性，通过与LMS算法的比较，验证了WO-LMS算法的优越性。     

时滞和滤波联合辨识的递推最小二乘算法

时滞和滤波联合辨识问题既是经典时滞估计问题的推广，又是自适应系统建模和时灌估计两方面的交叉。本文针对先滤波后时滞的系统模型，从方法上改进BOUDREAU＆KABAL提出的基于快速横向滤波器的递谁最小二乘滤波算法，使其时间复杂度由O［19ρ］下降为0［7ρ］，便于实时在线应用。我们以低通滤波器与线性时滞串联系统的辨识为例，表现该算法对变化时滞的跟踪能力及联合辨识性能。     

一种用于雷达量测的滤波算法

在雷达跟踪系统中,广泛使用混合坐标系下扩展卡尔曼滤波器（EKF）和转换卡尔曼滤波器（CMKF）两种算法,但是当目标距离较远时,这两种滤波器由于量测方程非线性的影响,误差较大,甚至导致滤波发射。本文计算了传统EKF滤波器引入的线性化误差,并在此基础上提出一种新的滤波算法（NDRKF）。计算机仿具结果表明,该算法减小了线性化误差的影响,较明显地改善了滤波顺的性能。     

一种高精度的BDS/INS紧耦合测姿算法

针对"北斗"卫星姿态测量系统在测姿过程中测姿精度和稳定性不高的问题,提出了"北斗"卫星导航系统(BDS)/惯性导航系统(INS)紧耦合姿态测量算法.该算法首先利用BDS观测量设计了BDS系统测姿误差模型;然后以INS状态误差方程为滤波系统状态方程,以载波相位为主要观测量设计了扩展卡尔曼滤波器,利用滤波器的输出实现对惯性导航测姿系统的辅助校正;最后采用静态测试、动态测试和遮挡测试验证该算法.该系统可以有效提高BDS测姿精度与输出频率,并且在静态条件下航向角测量精度可以达到0.15°.     

自适应滤波器的算法研究及DSP仿真实现

随着DSP技术的高速发展,人们对信号处理的实时性、准确性和灵活性的要求越来越高,DSP技术在信号处理中的地位也越来越重要.自适应滤波器是一种复杂的算法,设计它是为了在均衡信道,抵消回波,增强谱线,抑制噪声等方面有所应用.而自适应滤波器的实现主要采用最小均方误差算法完成.自适应算法通过调整滤波器系数来实现可以更好地跟踪信号的变化,最终实现自适应滤波.     

基于PSO和改进神经网络的图像滤波方法的研究

提出了一种基于改进BP神经网络和粒子群优化算法（PSO）的图像滤波方法。该方法利用对数最小均方误差函数（LNLS）代替BP神经网络传统的最小均方误差函数（LMS）,用来减小图像噪声对神经网络精度的影响;并将改进后的BP神经网络利用PSO算法优化,从而避免神经网络陷入局部极小值点,进一步提高神经网络滤波能力。实验结果表明,与传统滤波方法相比,该方法不仅能有效地滤除图像中的高斯噪声而且能很好地保护图像细节。     

回声消除基于NLMS改进算法的研究

为了提高VoIP的通信质量,减少回声干扰,对LMS算法、NLMS算法进行阐述,基于NLMS提出了一种运算量小并且提高收敛性能的改进的自适应滤波算法。通过在Matlab下的仿真研究和对误差曲线的分析,证明了该改进算法的收敛速度快,均方误差小。用改进的算法对语音回声信号进行消除,仿真得到消除回声后的信号效果明显,为IP电话中回声消除的自适应滤波问题提供了一个较好的算法。     

基于自适应滤波的压缩感知重构方法

自适应滤波框架中,滤波器的抽头系数可以利用特定的自适应算法达到近似维纳解,从而使滤波器的输出误差达到最小.将这个框架应用到压缩感知重构信号中,信号的稀疏系数等效为滤波器系数权值向量,从而可获得最佳的稀疏系数,以高概率重构信号.本文介绍了已有学者研究出的一种L0最小均方算法(L0-LMS),该算法中引入零引力项加快了权矢量向稀疏解收敛的速度,保证解的稀疏性.通过仿真可知,基于自适应滤波算法重构稀疏信号的性能较好,甚至优于压缩感知中常用的OMP算法.     

单站目标被动跟踪的滤波器方法研究

为了实现对单站目标的被动跟踪,分析并比较了扩展Kalman滤波器和粒子滤波器在非线性估计方面的性能,并且针对粒子滤波器存在的粒子退化现象,引入改进的重采样算法和基于无迹变换的滤波算法.仿真实验分别比较了几种滤波器在目标做匀速、匀加速、变加速情况下距离和速度滤波的均方根误差,结果表明粒子滤波器滤波性能优于扩展的Kalman滤波器,改进的重采样算法和基于无迹变换的粒子滤波器可以有效改善估计精度.     

Properties of FXLMS-Based Narrowband Active Noise Control With Online Secondary-Path Modeling

Rotating machines such as diesel engines, cutting machines, fans, motors, etc., generate sinusoidal noise signals that may be effectively reduced by narrowband active noise control (ANC) systems. In this paper, a typical filtered-X LMS (FXLMS) based narrowband ANC system equipped with an online secondary-path modeling subsystem is analyzed in detail. First, difference equations governing the dynamics of the FXLMS algorithm for secondary source synthesis and the LMS algorithm for secondary-path estimation are derived in terms of convergence in both mean and mean square. Steady-state expressions for mean-square error (MSE) as well as the residual noise power are then developed in closed form. Extensive simulations are performed to demonstrate the validity of the analytical results.     

基于sigmoid函数的Volterra自适应有源噪声对消器

该文介绍了一种新颖的非线性自适应有源噪声对消器－基于sigmoid函数的Volterra自适应有源噪声对消器，并采用输入信号和瞬时误差归一化的LMS自适应算法调整其系数。这种基于sigmoid函数的Volterra自适应有源噪声对消器具有参数少和便于实现的模块化结构等优点。仿真结果表明：这种基于sigmoid函数的Volterra自适应有源噪声对消系统具有良好的抗噪声性能。     

ELMS算法在噪声抵消中的应用

介绍一种改进的算法在噪声消除中的应用.算法利用第二级滤波器估计期望信号,然后在第一级滤波器中消除期望信号对算法误差的影响.仿真表明算法可以明显地降低算法的均方误差,得到很好的输出性能.     

A complex adaptive notch filter using modified gradient algorithm

A modified gradient algorithm is developed for improving the convergence speed of a first-order complex adaptive IIR notch filter, which is used for estimating an unknown frequency of a complex sinusoidal signal embedded in white Gaussian noise. The new cost function using new error criterion is presented and analyzed theoretically. The proposed technique can significantly improve the convergence speed as compared with a complex notch filter using plain gradient algorithm. The computer simulations are conducted to demonstrate the validity of the proposed complex adaptive notch filter.     

时变参数比例自适应滤波算法

在免提电话和视频会议系统中,自适应滤波器估计的回声路径通常是稀疏的.改进的比例归一化最小均方(IPNLMS)算法能够加快自适应滤波器在估计稀疏系统时的收敛速度,但与归一化最小均方(NLMS)算法相比,其稳态失调的波动性较大.为了解决这一问题,本文提出了一种时变参数IPNLMS(TV-IPNLMS)算法.该算法根据系统的均方误差(MSE)与噪声功率的比值,使用一个sigmoid函数来调整时变参数的值.该时变参数能够降低IPNLMS算法在滤波器到达稳态时的比例增益.仿真结果表明,时变参数方法能够降低IPNLMS算法稳态失调的波动性.该算法可用于回声消除、主动噪声控制等领域.     

A new adaptive recursive RLS-based fast-array IIR filter for active noise and vibration control systems

Infinite impulse response filters have not been used extensively in active noise and vibration control applications. The problems are mainly due to the multimodal error surface and instability of adaptive IIR filters used in such applications. Considering these, in this paper a new adaptive recursive RLS-based fast-array IIR filter for active noise and vibration control applications is proposed. At first an RLS-based adaptive IIR filter with computational complexity of order O(n²) is derived, and a sufficient condition for its stability is proposed by applying passivity theorem on the equivalent feedback representation of this adaptive algorithm. In the second step, to reduce the computational complexity of the algorithm to the order of O(n) as well as to improve its numerical stability, a fast array implementation of this adaptive IIR filter is derived. This is accomplished by extending the existing results of fast-array implementation of adaptive FIR filters to adaptive IIR filters. Comparison of the performance of the fast-array algorithm with that of Erikson’s FuLMS and SHARF algorithms confirms that the proposed algorithm has faster convergence rate and ability to reach a lower minimum mean square error which is of great importance in active noise and vibration control applications.     

Hybrid filtered error LMS algorithm: another alternative to filtered-x LMS

The filtered-error LMS (FELMS) algorithms are widely used in multi-input and multi-output control (MIMO) active noise control (ANC) systems as an alternative to the filtered-x LMS (FXLMS) algorithms to reduce the computational complexity and memory requirements. However, the available FELMS algorithms introduce significant delays in updating the adaptive filter coefficients that slow the convergence rate. In this paper, we introduce a novel algorithm called the hybrid filtered-error LMS algorithm (HFELMS) which, while still a form of the FELMS algorithm, allows users to have some freedom to construct the error filter that guarantees its convergence with a sufficiently small step size. Without increasing the computational complexity, the proposed algorithm can improve the control system performance in one of several ways: 1) increasing the convergence rate without extra computation cost; 2) reducing the remaining noise mean square error (MSE); or 3) shaping the excess noise power. Simulation results show the effectiveness of the proposed method.     

一种窄带主动噪声控制系统次声学路径建模技术

本文针对最大长度序列相关(Maximal Length sequence Correlation，MLc)建模技术在窄带主动噪声控制系统次声学路径建模的应用中，系统性能易受窄带信号影响这一不足，提出了一种改进的MLC(MMLC)次声学路径建模技术。具体地说就是采用一个自适应预测滤波器来预测和消除MLC技术中的窄带干扰，并用一个补偿滤波器来修正由预测误差滤波器引起的训练信号成分失真。计算机仿真表明，MMLC算法能有效克服窄带主动噪声控制系统次声学路径建模的窄带信号影响，具有较高的建模精度。     

Feedback active noise control based on transform-domain forward–backward LMS predictor

In this paper, a new feedback active noise control (FBANC) system based on the transform-domain forward–backward LMS (TFBLMS) predictor has been proposed. The new ANC system employs the TFBLMS predictor for its main-path (MP) predictor as well as for the noise canceller. To overcome the ill effect of the primary noise field, which acts as an observation noise for the secondary-path (SP) identification, the noise canceller is used. As the main-path predictor is based on the TFBLMS, its convergence rate improves due to its input orthogonalization. Further, its FBLMS nature reduces misadjustment. The use of TFBLMS predictor for noise canceller also gives a good prediction of primary noise at a faster rate, enabling improved SP identification. This improved SP identification indirectly aids the MP predictor to achieve an improved performance. A new filtered-x LMS structure has been proposed to realize the new MP predictor to accommodate the TFBLMS algorithm. The TFBLMS algorithm is applied directly to the noise canceller for SP identification. The proposed new ANC system has been found to have a significantly better noise reduction (by 14.6 dB) over the FBANC system based on tapped delay line time-domain FBLMS algorithm.     

基于卡尔曼滤波算法船舶AIS轨迹估计研究

AIS是集船舶导航、避碰、海事监管于一体的现代化无线电系统,在水上交通运输领域里发挥着重要作用。为了弥补AIS数据堵塞等原因导致更新数据不及时,造成船舶轨迹的不准确或者误差较大的问题,提出利用对卡尔曼滤波算法进行适当的修改,引入系统噪声和测量噪声,利用AIS船舶观测节点数据对系统状态做最小二乘法估计,对船舶运动轨迹进行平滑和预测,并对不同时间段可靠性误差进行分析,能够比较正确地估计出船舶轨迹。