针对风洞试验模型系统辨识不准确的问题,利用自适应LMS(least mean square)滤波器模型对跨声速风洞模型进行系统辨识。由于实测信号中存在多模态耦合,为了提高系统辨识精准度,首先对输入输出信号作了FRF(frequency response analysis)...针对风洞试验模型系统辨识不准确的问题,利用自适应LMS(least mean square)滤波器模型对跨声速风洞模型进行系统辨识。由于实测信号中存在多模态耦合,为了提高系统辨识精准度,首先对输入输出信号作了FRF(frequency response analysis)分析得到试验模型俯仰方向前两阶模态,其次利用快速Fourier变换进行模态解耦,接着利用自适应LMS滤波器模型、传递函数模型、多项式模型对俯仰方向单模态进行系统辨识,最后得到了基于自适应LMS滤波器模型的俯仰方向一阶、二阶模态滤波器系数。通过对比不同数学模型的输出与输入之间的相关系数和均方误差及辨识结果,表明自适应LMS滤波器模型具有更高的系统辨识精准度和更简洁的数学模型结构。为后续风洞试验模型振动主动控制计算法的设计提供有力支撑。展开更多
Filter bank multicarrier quadrature amplitude modulation(FBMC-QAM)will encounter inter-ference and noise during the process of channel transmission.In order to suppress the interference in the communication system,cha...Filter bank multicarrier quadrature amplitude modulation(FBMC-QAM)will encounter inter-ference and noise during the process of channel transmission.In order to suppress the interference in the communication system,channel equalization is carried out at the receiver.Given that the con-ventional least mean square(LMS)equilibrium algorithm usually suffer from drawbacks such as the inability to converge quickly in large step sizes and poor stability in small step sizes when searching for optimal weights,in this paper,a design scheme for adaptive equalization with dynamic step size LMS optimization is proposed,which can further improve the convergence and error stability of the algorithm by calling the Sigmoid function and introducing three new parameters to control the range of step size values,adjust the steepness of step size,and reduce steady-state errors in small step sta-ges.Theoretical analysis and simulation results demonstrate that compared with the conventional LMS algorithm and the neural network-based residual deep neural network(Res-DNN)algorithm,the adopted dynamic step size LMS optimization scheme can not only obtain faster convergence speed,but also get smaller error values in the signal recovery process,thereby achieving better bit error rate(BER)performance.展开更多
为解决自适应最小均方误差(least mean squares,LMS)滤波算法难以平衡稳态误差和收敛速度的问题,提出了基于对称非线性函数的变步长LMS自适应滤波算法。通过自变量取绝对值、叠加非线性拉伸量改进Sig-moid函数,构造一个对称非线性函数...为解决自适应最小均方误差(least mean squares,LMS)滤波算法难以平衡稳态误差和收敛速度的问题,提出了基于对称非线性函数的变步长LMS自适应滤波算法。通过自变量取绝对值、叠加非线性拉伸量改进Sig-moid函数,构造一个对称非线性函数用于刻画步长因子与稳态误差的非线性关系。该对称非线性函数具有能够根据误差动态调整步长、更快达到收敛状态的特点。根据构造的对称非线性函数和输入信号功率生成归一化变步长因子,解决噪声逐级放大的问题,进一步提高算法的滤波效果同时,加速收敛。实验表明:该算法在低信噪比、信噪比变化、信号频率变化、滤波器阶数变化、延迟采样点数变化条件下均具有更好的滤波效果、更优的稳定性和更快的收敛速度。展开更多
Random vibration control is aimed at reproducing the power spectral density (PSD) at specified control points. The classical frequency-spectrum equalization algorithm needs to compute the average of the multiple fre...Random vibration control is aimed at reproducing the power spectral density (PSD) at specified control points. The classical frequency-spectrum equalization algorithm needs to compute the average of the multiple frequency response functions (FRFs), which lengthens the control loop time in the equalization process. Likewise, the feedback control algorithm has a very slow convergence rate due to the small value of the feedback gain parameter to ensure stability of the system. To overcome these limitations, an adaptive inverse control of random vibrations based on the filtered-X least mean-square (LMS) algorithm is proposed. Furthermore, according to the description and iteration characteristics of random vibration tests in the frequency domain, the frequency domain LMS algorithm is adopted to refine the inverse characteristics of the FRF instead of the traditional time domain LMS algorithm. This inverse characteristic, which is called the impedance function of the system under control, is used to update the drive PSD directly. The test results indicated that in addition to successfully avoiding the instability problem that occurs during the iteration process, the adaptive control strategy minimizes the amount of time needed to obtain a short control loop and achieve equalization.展开更多
为了避免单个滤波器在收敛速度与稳态误差上相互制约,从而导致系统性能降低的问题,本文采用凸组合最小均方算法(Combined Least Mean Square,CLMS),将快速滤波器和慢速滤波器并联使用,同时为进一步改善CLMS算法的性能,对已有的变步长凸...为了避免单个滤波器在收敛速度与稳态误差上相互制约,从而导致系统性能降低的问题,本文采用凸组合最小均方算法(Combined Least Mean Square,CLMS),将快速滤波器和慢速滤波器并联使用,同时为进一步改善CLMS算法的性能,对已有的变步长凸组合最小均方算法(Variable Step-size Convex Combination of LMS,VSCLMS)做出改进,提出了一种新的VSCLMS算法.在该算法中,对快速滤波器选用以最小均方权值偏差(Minimization of Mean Square Weight Error,MMSWE)为准则的按步分析的变步长滤波器;对慢速滤波器采用以稳态最小均方误差(Least Mean Square,LMS)为准则的固定步长滤波器.通过理论分析与仿真实验表明,该算法能够在噪声、时变以及非平稳的环境下保持较好的随动性能,且在各个阶段均保持良好的收敛性,与传统的CLMS、VSCLMS算法相比,不仅具有更快的收敛速度,而且拥有稳定的均方性能和较优的跟踪性能,为自适应滤波算法的研究提供了一条可行途径.展开更多
文摘针对风洞试验模型系统辨识不准确的问题,利用自适应LMS(least mean square)滤波器模型对跨声速风洞模型进行系统辨识。由于实测信号中存在多模态耦合,为了提高系统辨识精准度,首先对输入输出信号作了FRF(frequency response analysis)分析得到试验模型俯仰方向前两阶模态,其次利用快速Fourier变换进行模态解耦,接着利用自适应LMS滤波器模型、传递函数模型、多项式模型对俯仰方向单模态进行系统辨识,最后得到了基于自适应LMS滤波器模型的俯仰方向一阶、二阶模态滤波器系数。通过对比不同数学模型的输出与输入之间的相关系数和均方误差及辨识结果,表明自适应LMS滤波器模型具有更高的系统辨识精准度和更简洁的数学模型结构。为后续风洞试验模型振动主动控制计算法的设计提供有力支撑。
基金the National Natural Science Foundation of China(No.61601296,61701295)the Science and Technology Innovation Action Plan Project of Shanghai Science and Technology Commission(No.20511103500)the Talent Program of Shanghai University of Engineering Science(No.2018RC43).
文摘Filter bank multicarrier quadrature amplitude modulation(FBMC-QAM)will encounter inter-ference and noise during the process of channel transmission.In order to suppress the interference in the communication system,channel equalization is carried out at the receiver.Given that the con-ventional least mean square(LMS)equilibrium algorithm usually suffer from drawbacks such as the inability to converge quickly in large step sizes and poor stability in small step sizes when searching for optimal weights,in this paper,a design scheme for adaptive equalization with dynamic step size LMS optimization is proposed,which can further improve the convergence and error stability of the algorithm by calling the Sigmoid function and introducing three new parameters to control the range of step size values,adjust the steepness of step size,and reduce steady-state errors in small step sta-ges.Theoretical analysis and simulation results demonstrate that compared with the conventional LMS algorithm and the neural network-based residual deep neural network(Res-DNN)algorithm,the adopted dynamic step size LMS optimization scheme can not only obtain faster convergence speed,but also get smaller error values in the signal recovery process,thereby achieving better bit error rate(BER)performance.
文摘为解决自适应最小均方误差(least mean squares,LMS)滤波算法难以平衡稳态误差和收敛速度的问题,提出了基于对称非线性函数的变步长LMS自适应滤波算法。通过自变量取绝对值、叠加非线性拉伸量改进Sig-moid函数,构造一个对称非线性函数用于刻画步长因子与稳态误差的非线性关系。该对称非线性函数具有能够根据误差动态调整步长、更快达到收敛状态的特点。根据构造的对称非线性函数和输入信号功率生成归一化变步长因子,解决噪声逐级放大的问题,进一步提高算法的滤波效果同时,加速收敛。实验表明:该算法在低信噪比、信噪比变化、信号频率变化、滤波器阶数变化、延迟采样点数变化条件下均具有更好的滤波效果、更优的稳定性和更快的收敛速度。
基金Program for New Century Excellent Talents in Universities Under Grant No.NCET-04-0325
文摘Random vibration control is aimed at reproducing the power spectral density (PSD) at specified control points. The classical frequency-spectrum equalization algorithm needs to compute the average of the multiple frequency response functions (FRFs), which lengthens the control loop time in the equalization process. Likewise, the feedback control algorithm has a very slow convergence rate due to the small value of the feedback gain parameter to ensure stability of the system. To overcome these limitations, an adaptive inverse control of random vibrations based on the filtered-X least mean-square (LMS) algorithm is proposed. Furthermore, according to the description and iteration characteristics of random vibration tests in the frequency domain, the frequency domain LMS algorithm is adopted to refine the inverse characteristics of the FRF instead of the traditional time domain LMS algorithm. This inverse characteristic, which is called the impedance function of the system under control, is used to update the drive PSD directly. The test results indicated that in addition to successfully avoiding the instability problem that occurs during the iteration process, the adaptive control strategy minimizes the amount of time needed to obtain a short control loop and achieve equalization.
文摘为了避免单个滤波器在收敛速度与稳态误差上相互制约,从而导致系统性能降低的问题,本文采用凸组合最小均方算法(Combined Least Mean Square,CLMS),将快速滤波器和慢速滤波器并联使用,同时为进一步改善CLMS算法的性能,对已有的变步长凸组合最小均方算法(Variable Step-size Convex Combination of LMS,VSCLMS)做出改进,提出了一种新的VSCLMS算法.在该算法中,对快速滤波器选用以最小均方权值偏差(Minimization of Mean Square Weight Error,MMSWE)为准则的按步分析的变步长滤波器;对慢速滤波器采用以稳态最小均方误差(Least Mean Square,LMS)为准则的固定步长滤波器.通过理论分析与仿真实验表明,该算法能够在噪声、时变以及非平稳的环境下保持较好的随动性能,且在各个阶段均保持良好的收敛性,与传统的CLMS、VSCLMS算法相比,不仅具有更快的收敛速度,而且拥有稳定的均方性能和较优的跟踪性能,为自适应滤波算法的研究提供了一条可行途径.