一类 Wiener 非线性时变系统的迭代学习辨识

针对Wiener非线性时变系统的参数辨识问题,该文提出一种基于重复轴的迭代学习算法来实现对时变甚至突变参数的估计。文中将维纳系统输出非线性部分的反函数进行多项式展开,进而构造了回归模型,未知参数及中间变量用其估计替代,分别给出了采用迭代学习梯度算法和迭代学习最小二乘算法实现时变参数辨识的方法。仿真结果表明,与带遗忘因子的递推算法和迭代学习梯度算法相比,迭代学习最小二乘算法更具有参数估计收敛速度快,辨识精度高,系统输出误差小等优势,验证了所提学习算法的有效性。     

一类Wiener非线性时变系统的迭代学习辨识

针对Wiener非线性时变系统的参数辨识问题,该文提出一种基于重复轴的迭代学习算法来实现对时变甚至突变参数的估计.文中将维纳系统输出非线性部分的反函数进行多项式展开,进而构造了回归模型,未知参数及中间变量用其估计替代,分别给出了采用迭代学习梯度算法和迭代学习最小二乘算法实现时变参数辨识的方法.仿真结果表明,与带遗忘因子的递推算法和迭代学习梯度算法相比,迭代学习最小二乘算法更具有参数估计收敛速度快,辨识精度高,系统输出误差小等优势,验证了所提学习算法的有效性.     

变遗忘因子递推最小二乘Hammerstein系统辨识算法

摘.要:本论文研究了单输入单输出非线性Hammerstein系统的辨识问题,提出了一种具有变遗忘因子的递推最小二乘算法.由于Hammerstein系统模型的非线性特征,传统的递推最小二乘算法无法直接用来解决该系统的辨识问题.为此,论文将Hammerstein系统参数进行了映射变换,使得变换后的系统参数与Hammerst...     

带遗忘因子的限定记忆辨识算法

基于最小二乘辨识原理,针对快时变系统,提出了一种将遗忘因子算法与限定记忆算法相结合的辨识算法,该方法在固定记忆长度中加入遗忘因子,并根据估计参数的变化率来在线调节遗忘因子的大小,提高了时变系统的辨识精度,数值仿真结果表明该方法有较好的辨识效果。     

基于自调整器的CDMA系统盲自适应干扰抑制

本文给出一种能同时抑制DS-CDMA系统多址干扰(MAI)和窄带干扰(NBI)的盲自适应算法.此方法基于遗忘因子具有自调整器的迭代最小二乘算法(SR-RLS),根据系统的变化自动调整遗忘因子,当系统趋于静态时,遗忘因子趋于1,以提高稳态精度,在动态系统中,遗忘因子减小,使算法能有效的跟踪系统参数.与其它的迭代最小二乘相比,具有较小的稳态误差和良好的动态跟踪能力.文章从理论上分析了算法的收敛性.最后,对算法在静态环境和动态环境中的性能分别进行了仿真分析.     

一种新的用于Hammerstein预失真器的自适应结构

针对目前的自适应预失真结构不利于高效的最小二乘算法直接对Hammerstein预失真器参数进行更新的问题,该文提出了一种新的自适应预失真结构。应用该结构可以得到Hammerstein预失真器中两个子系统的误差,因此可使用高效的最小二乘算法直接对Hammerstein预失真器进行自适应更新,避免了结构误差以及子系统误差不精确对预失真器性能的影响。仿真结果表明:该文提出的自适应结构可使Hammerstein预失真器快速高效地补偿带记忆效应功率放大器的非线性失真。     

MIMO-OFDM系统中基于迭代MAP算法的RLS信道估计

该文提出了一种适用于MIMO-OFDM系统的迭代最大后验概率(Iterative-MAP)信道估计算法。接收机利用MAP译码算法中的信息位和校验位软信息,经过非线性映射将信息反馈至信道估计模块,采用递归最小二乘(RLS)自适应滤波算法对信道时变状态参数进行跟踪,提高了信道估计的精度。仿真结果表明,该方法与最小二乘(LS)算法相比,估计的均方误差(MSE)和误帧率(FER)性能都有较大改善。     

基于混沌压缩感知的稀疏时变信号在线估计

混沌压缩感知是一种利用混沌系统实现非线性测量的压缩感知理论。针对稀疏时变信号的混沌压缩感知,该文提出稀疏时变信号的在线估计架构,构建一种递归最小二乘准则下的稀疏约束目标函数;通过利用迭代加权非线性最小二乘算法求解目标函数最小化问题,实现稀疏时变信号的参数估计。以Henon混沌系统为例仿真分析了频域时变稀疏信号的估计性能,数值模拟证明了该方法的有效性。     

非凸投影自适应Hammerstein滤波算法

该文研究了Hammerstein系统参数辨识和非线性系统预测问题,提出一种基于非凸投影的自适应滤波算法。论文将问题归结为具有非凸可行域的约束优化问题,并建立了基于交替方向乘子法(ADMM)和递归最小二乘相结合的算法框架。在该算法框架下,非凸约束优化问题的全局最优解可通过岭回归和欧几里得(Euclid)投影循环计算得到。将提出的算法分别应用于Hammerstein系统的参数辨识、非线性未知系统预测以及非线性声学回声消除,并进行仿真实验,结果显示所提算法具有较好的收敛性和稳定性,能够得到较准确的辨识和预测效果。     

基于BSNN-ARX的光伏逆变器模型辨识

光伏逆变器是光伏并网系统的核心部件,将基于Hammerstein模型的非线性系统辨识方法引入到光伏并网逆变器的建模中,把单相光伏并网逆变器视为双输入单输出的非线性黑箱系统。在Hammerstein模型的静态非线性环节采用B样条神经网络,动态线性环节采用ARX模型,同时采用基于误差学习准则和最小二乘递归准则的自适应学习方法。实验测试结果表明,提出的BSNN-ARX光伏逆变器模型辨识方法可以对不同天气条件下的逆变器输出功率进行高精度的辨识,从而为并网逆变器的建模提供一种有效途径。     

Decomposition Based Newton Iterative Identification Method for a Hammerstein Nonlinear FIR System with ARMA Noise

This paper derives a Newton iterative algorithm for identifying a Hammerstein nonlinear FIR system with ARMA noise (i.e., Hammerstein nonlinear controlled autoregressive moving average system). This method decomposes a Hammerstein nonlinear system into two subsystems using the hierarchical identification principle, estimating the parameters of the system directly without using the over-parameterization method. The simulation results show that the proposed algorithm is effective.     

Identification of certain time-varying nonlinear Wiener andHammerstein systems

The problem of identification and tracking of time-varying nonlinear systems is addressed. In particular, the Wiener system that consists of a dynamic time-varying linear part followed by a fixed nonlinearity and the Hammerstein system in which the order of these two blocks is reversed are studied. The extended Kalman filter (EKF) algorithm is applied. It is also shown that this algorithm can be reformulated in terms of a nonlinear minimization problem with a quadratic inequality constraint in order to ensure exponential stability, resulting in the algorithm CEKF. As indicated by means of numerical examples, this latter algorithm is less sensitive to the chosen initialization than the EKF. The proposed algorithms depend on certain second-order statistics that may be unknown in a typical scenario. A method for estimation of these quantities is proposed. It is demonstrated that the suggested algorithms can be successfully applied to the problem of acoustic echo cancelation     

基于动态加权LS-SVM的网络流量混沌预测

网络流量是具有复杂非线性、不确定时变性的混沌时间序列.为提高标准最小二乘支持向量机的预测精度与自适应性,提出一种基于动态加权最小二乘支持向量机的网络流量混沌预测方法.该方法在标准LS-SVM回归机的训练样本误差设置时间权,增强对非线性样本的逼近能力.然后结合滚动窗与迭代求逆法实现模型动态在线校正,进而克服网络变化时的累积误差.仿真实验结果表明,相对常规LS-SVM,该模型能降低预测误差、减少计算时间,实现高精度实时混沌流量估计.     

A Multi-innovation Recursive Least Squares Algorithm with a Forgetting Factor for Hammerstein CAR Systems with Backlash

This study addresses the identification of Hammerstein CAR systems with backlash, where the nonlinear backlash is described as one regression identification model using a two switching function mathematical model. In such a case, the Hammerstein CAR systems with backlash can be transformed into a piecewise linearized model. Then, a novel multi-innovation recursive least squares algorithm with a forgetting factor is applied to estimate the parameters of the proposed model. Finally, numerical examples are presented to test the performance of the proposed algorithm.     

Identification of Hammerstein systems with time-varying piecewise-linear characteristics

This brief deals with the recursive parameter identification of Hammerstein type nonlinear dynamic systems with time-varying piecewise-linear characteristics. A special form of the Hammerstein model, which is linear in parameters, is incorporated into the recursive least squares identification scheme supplemented with the estimation of model internal variables. This enables online estimation of the linear block parameters, the coefficients determining the partition of nonlinearity subdomains and the corresponding linear segment slopes. An illustrative example is included.     

Least Squares Identification for Hammerstein Multi-input Multi-output Systems Based on the Key-Term Separation Technique

System modeling and parameter estimation are basic for system analysis and controller design. This paper considers the parameter identification problem of a Hammerstein multi-input multi-output (H-MIMO) system. In order to avoid the product terms in the identification model, we derive a pseudo-linear identification model of the H-MIMO system through separating a key term from the output equation of the system and present a hierarchical generalized least squares (LS) algorithm for estimating the parameters of the system. Moreover, we present a new LS algorithm to reduce the computational burden. The proposed algorithms are simple in principle and can achieve a higher computational efficiency than the over-parameterization-based LS estimation algorithm. Finally, we test the proposed algorithms by the simulation example and show their effectiveness.     

Iterative least squares estimators in nonlinear image restoration

The concept of iterative least squares estimation as applied to nonlinear image restoration is considered. Regarding the convergence analysis of nonlinear iterative algorithms, the potential of the global convergence theorem (GCT) is explored. The theoretical analysis is performed on a general class of nonlinear algorithms, which defines a signal-dependent linear mapping of the residual. The descent properties of two normed functions are considered. Furthermore, a procedure for the selection of the iteration parameter is introduced. The steepest descent (SD) iterative approach for the solution of the least squares optimization problem is introduced. The convergence properties of the particular algorithm are readily derived on the basis of the generalized analysis and the GCT. The factors that affect the convergence rate of the SD algorithm are thoroughly studied. In the case of the SD algorithm, structural modifications are proposed, and two hybrid-SD algorithms attain convergence in a more uniform fashion with respect to their entries. In general, the algorithms achieve larger convergence rates than the conventional SD technique     

Recursive Parameter Estimation Algorithms and Convergence for a Class of Nonlinear Systems with Colored Noise

Parameter estimation is important for controller design of linear systems and nonlinear systems. The parameters of the systems can be estimated through some identification algorithms. This paper presents a recursive generalized extended least squares algorithm and a generalized extended stochastic gradient (GESG) algorithm for identifying the parameters of a class of nonlinear systems. Furthermore, a multi-innovation GESG algorithm is derived to improve the estimation accuracy. The simulation example is provided to test the effectiveness of the proposed algorithms.