论文部分内容阅读
利用双曲正切函数对凸组合最小均方(Convex Combination of Least Mean Square,CLMS)算法进行了改进,提出了Th-VSCLMS(Tanh Variable Step Convex Combined Least Mean Square)算法。理论分析表明:改进算法的性能优于普通的变步长算法。在不同信噪比下对VSLMS(Variable Step Least Mean Square)、CLMS、Th-VSCLMS算法进行了仿真对比分析,结果表明:Th-VSCLMS算法的收敛性、稳态性、抗扰动性、低信噪比的滤波效果、计算量均得到较好的改善,且其凸组合滤波结构优于单一滤波结构。
The Consex Combination of Least Mean Square (CLMS) algorithm is improved by using the hyperbolic tangent function, and a ThhVCLCL (Tanh Variable Step Convex Combined Least Mean Square) algorithm is proposed. The theoretical analysis shows that the performance of the improved algorithm is better than the ordinary variable-step algorithm. The comparison and analysis of VSLMS (Variable Step Least Mean Square), CLMS and Th-VSCLMS algorithms under different signal-to-noise ratios show that the convergence, steady state, anti-disturbance and low signal-noise Compared with the filtering effect, the calculated amount has been better improved, and its convex combination filter structure is better than a single filter structure.