A type of algorithms of variable step-size LMS adaptive filtering is discussed and presented another new non-linear functional relationship between the step-size and the error signalk, which improves the algorithm perform anee. The new algorithm has better MSE convergence properties and smaller stable-state maladjustment. It also keeps excellent tracking ability at low noise. Finally, simulation results show the algorithm is wuperior to the former algorithms in performance.%为了提高LMS自适应算法的性能,在对传统LMS算法、变步长LMS算法及其改进算法分析的基础上,提出了一种新的步长与误差信号之间的函数关系.既能保证均方误差收敛、有较小的失凋,又能在算法达到稳态时,在低信噪比环境下有很好的跟踪性能,并且不易受噪声影响.理论分析和计算机仿真结果均表明该算法优于传统算法.
展开▼