针对手机终端语音降噪处理系统中,广泛使用计算复杂度较小的固定步长LMS(Least Mean Square)自适应滤波算法。本文提出一种通过建立步长因子μ(n)和误差信号e(n)非线性函数关系,进而把固定步长LMS自适应算法改进为新的变步长自适应滤波...针对手机终端语音降噪处理系统中,广泛使用计算复杂度较小的固定步长LMS(Least Mean Square)自适应滤波算法。本文提出一种通过建立步长因子μ(n)和误差信号e(n)非线性函数关系,进而把固定步长LMS自适应算法改进为新的变步长自适应滤波算法,较好地解决了固定步长因子无法解决收敛速度和稳态误差之间相矛盾的不足。最后,给出了改进型算法在ZSP800核数字信号处理器(DSP)上的实现方法。展开更多
We study iterative processes of stochastic approximation for finding fixed points of weakly contractive and nonexpansive operators in Hilbert spaces under the condition that operators are given with random errors. We ...We study iterative processes of stochastic approximation for finding fixed points of weakly contractive and nonexpansive operators in Hilbert spaces under the condition that operators are given with random errors. We prove mean square convergence and convergence almost sure (a.s.) of iterative approximations and establish both asymptotic and nonasymptotic estimates of the convergence rate in degenerate and non-degenerate cases. Previously the stochastic approximation algorithms were studied mainly for optimization problems.展开更多
针对解决LMS(Least Mean Square)自适应滤波算法收敛速度及未知系统时变跟踪速度与稳态误差的矛盾,改进步长因子μ(n)与误差信号e(n)的非线性映射关系,提出一种新的变步长LMS算法。执行该算法时系统初始阶段或未知系统时变阶段步长自动...针对解决LMS(Least Mean Square)自适应滤波算法收敛速度及未知系统时变跟踪速度与稳态误差的矛盾,改进步长因子μ(n)与误差信号e(n)的非线性映射关系,提出一种新的变步长LMS算法。执行该算法时系统初始阶段或未知系统时变阶段步长自动增大,而稳态时步长缓慢变小,提高了收敛速度和时变跟踪能力,克服了稳态误差偏大的缺点。理论分析及实验结果表明,新算法的收敛、跟踪速度及稳态误差性能均优于现有常见的几种LMS算法。展开更多
The goal of computational science is to develop models that predict phenomena observed in nature. However, these models are often based on parameters that are uncertain. In recent decades, main numerical methods for s...The goal of computational science is to develop models that predict phenomena observed in nature. However, these models are often based on parameters that are uncertain. In recent decades, main numerical methods for solving SPDEs have been used such as, finite difference and finite element schemes [1]-[5]. Also, some practical techniques like the method of lines for boundary value problems have been applied to the linear stochastic partial differential equations, and the outcomes of these approaches have been experimented numerically [7]. In [8]-[10], the author discussed mean square convergent finite difference method for solving some random partial differential equations. Random numerical techniques for both ordinary and partial random differential equations are treated in [4] [10]. As regards applications using explicit analytic solutions or numerical methods, a few results may be found in [5] [6] [11]. This article focuses on solving random heat equation by using Crank-Nicol- son technique under mean square sense and it is organized as follows. In Section 2, the mean square calculus preliminaries that will be required throughout the paper are presented. In Section 3, the Crank-Nicolson scheme for solving the random heat equation is presented. In Section 4, some case studies are showed. Short conclusions are cleared in the end section.展开更多
文摘针对手机终端语音降噪处理系统中,广泛使用计算复杂度较小的固定步长LMS(Least Mean Square)自适应滤波算法。本文提出一种通过建立步长因子μ(n)和误差信号e(n)非线性函数关系,进而把固定步长LMS自适应算法改进为新的变步长自适应滤波算法,较好地解决了固定步长因子无法解决收敛速度和稳态误差之间相矛盾的不足。最后,给出了改进型算法在ZSP800核数字信号处理器(DSP)上的实现方法。
文摘We study iterative processes of stochastic approximation for finding fixed points of weakly contractive and nonexpansive operators in Hilbert spaces under the condition that operators are given with random errors. We prove mean square convergence and convergence almost sure (a.s.) of iterative approximations and establish both asymptotic and nonasymptotic estimates of the convergence rate in degenerate and non-degenerate cases. Previously the stochastic approximation algorithms were studied mainly for optimization problems.
文摘针对解决LMS(Least Mean Square)自适应滤波算法收敛速度及未知系统时变跟踪速度与稳态误差的矛盾,改进步长因子μ(n)与误差信号e(n)的非线性映射关系,提出一种新的变步长LMS算法。执行该算法时系统初始阶段或未知系统时变阶段步长自动增大,而稳态时步长缓慢变小,提高了收敛速度和时变跟踪能力,克服了稳态误差偏大的缺点。理论分析及实验结果表明,新算法的收敛、跟踪速度及稳态误差性能均优于现有常见的几种LMS算法。
文摘The goal of computational science is to develop models that predict phenomena observed in nature. However, these models are often based on parameters that are uncertain. In recent decades, main numerical methods for solving SPDEs have been used such as, finite difference and finite element schemes [1]-[5]. Also, some practical techniques like the method of lines for boundary value problems have been applied to the linear stochastic partial differential equations, and the outcomes of these approaches have been experimented numerically [7]. In [8]-[10], the author discussed mean square convergent finite difference method for solving some random partial differential equations. Random numerical techniques for both ordinary and partial random differential equations are treated in [4] [10]. As regards applications using explicit analytic solutions or numerical methods, a few results may be found in [5] [6] [11]. This article focuses on solving random heat equation by using Crank-Nicol- son technique under mean square sense and it is organized as follows. In Section 2, the mean square calculus preliminaries that will be required throughout the paper are presented. In Section 3, the Crank-Nicolson scheme for solving the random heat equation is presented. In Section 4, some case studies are showed. Short conclusions are cleared in the end section.
