一种新的变步长LMS自适应算法及其在自适应噪声对消中的应用

许多时变步长(VSS)自适应算法已经提出用来完善标准LMS算法的性能,但实验表明这些算法容易受噪声干扰.本文介绍了一种新的变步长LMS自适应算法,这种算法保证了较小的失调,同时使算法在自适应初始阶段有较快的收敛速度.该算法的优越性在于它不受已经存在的不相关噪声的干扰.本文对该算法的收敛性和稳定性进行了分析,并将该算法应用于自适应噪声对消的仿真实验中,给出了计算机的仿真结果.

The New Variable Step-Size LMS-Type Algorithm and It's Application duringAdaptive Noise Cancellation

A number of time_varing size_algorithm have been proposed to enhance the performance of the cohventional LMS algorithm.But experimentation,with those algorithms indicates that their pertormance is highly sensitive to the noise disturbance .This paper present a new variable step_size LMS_type algorithm providing fast convergence at early stages of adaptation,while ensuring.small final misadjustment. The performance of the algorithm is not,affected by existing uncorrected noise disturbances..An analysis of convergence and steady_state performance of the algorithm is provided,and apply the algorithm in the .simulation of adaptive noise cancellation.Finally illustheive comparer simulation is given.