摘要
传统的最小均方误差(LMS)算法难以同时获取较快的收敛速度和较小的稳态误差,而变步长LMS算法可获得二者之间的平衡。对已有的一些变步长LMS算法进行了分析,在变系数步长(VFSS)算法的基础上,引入输入信号因子,并建立步长因子与误差信号之间新的非线性函数关系,提出一种改进的变步长LMS算法,该算法不仅继承了VFSS算法在低信噪比环境下抗噪声性能好的特点,而且能够快速跟踪系统的变化,仿真结果表明改进算法的性能优于现有算法。
The conventional LMS algorithm has difficulty in gaining the fast convergence speed and low steady state error at the same time.It can be overcome by means of a variable step size LMS algorithm.Some variable step size LMS algorithms in literature are analyzed.Based on VFSS algorithm,an improved variable step size LMS algorithm is proposed which introduces the input signal factor and establishes a new nonlinear function between the step size and the error signal.The algorithm not only maintains the advantage of VFSS algorithm that has good antinoise property under low SNR environment,but also can track changes fast in system.Simulation results show that the improved algorithm performance is better than that of former algorithms.
出处
《无线电通信技术》
2010年第4期13-15,共3页
Radio Communications Technology
关键词
LMS算法
变步长
收敛速度
抗噪声
LMS algorithm
variable step size
convergence speed
antinoise
