基于洛伦兹函数的变步长凸组合最小均方算法

为进一步减小收敛速率与稳态误差之间的矛盾,改善自适应滤波算法,利用改进的Lorentzian函数提出了一种新的变步长凸组合最小均方(new variable step-size convex combination of least mean square,NVS-CLMS)算法,该算法既有效提高了收敛速率又具备很好的抗干扰能力。同时,为了克服CLMS算法停滞等待的弊端,采用了瞬时转移结构;另外,在参数的迭代公式中使用sign函数进行优化以降低运算量。仿真结果证明该算法与CLMS、VS-CLMS相比,在不同的仿真环境中均能表现出良好的均方特性和跟踪特性。

Low computational complexity variable step-size CLMS algorithm based on Lorentzian function

In order to avoid the conflict between convergence speed and stable state error, and improve the adaptive filter algorithm, a new variable step-size convex combination of least mean square (NVS-CLMS) algorithm is proposed by using the improved Lorentzian function. The new algorithm effectively improves convergence rate and has good anti-interference performance. Furthermore, the instantaneous transfer scheme is utilized to overcome disadvantages of CLMS and the sign function in iterative of the parameter can also reduce the computational complexity. Theoretical analysis and simulation results show that under different environments, the proposed algorithm, compared with the CLMS and VS-CLMS algorithms, not only has a superior capability of tracking, but also can maintain a better convergence.