摘要
为解决传统最小均方自适应滤波算法联合参数λ(n)运算量大、收敛速度慢的问题,提出一种基于修正的反正切函数的凸组合最小均方滤波算法,并应用Matlab软件对不同信噪比算法进行仿真比较。结果表明:该算法在保证运算量的前提下,能够加快算法的收敛速度及减小其稳态误差。反正切函数联合参数的凸组合最小均方滤波算法具有更好的滤波性能。
Aimed at a solution to complex computation and poor convergence performance of the convex parametric of λ(n) in the least-mean-square(LMS) algorithm,this paper proposes a new CLMS algorithm based on improved arc-tangent function and introduces simulation comparison of algorithms with different SNR by using MATLAB math software.The simulation indicates that this new algorithm,able to quicken the convergence and reduce steady state error,along with assuring less computation,demonstrates a better filtering performance.
出处
《黑龙江科技学院学报》
CAS
2012年第6期608-612,共5页
Journal of Heilongjiang Institute of Science and Technology
关键词
凸组合
自适应滤波
反正切函数
convex combination
adaptive filter
arc-tangent function