在分析现有基于不同误差准则的最小四次均方算法和最小平均绝对偏差算法的基础上,提出一种可抑制多模噪声的归一化混合范数滤波算法。采用引入混合参数的代价函数调节混合参数得到最佳的滤波算法。建立步长参数随输入信噪比变化的非线性函数关系。仿真结果表明,相对于高斯环境下的滤波,该算法在多模噪声环境下的鲁棒性得到显著提高,能有效地提取信号。 Based on the analysis of the least square mean square algorithm and the minimum average absolute deviation algorithm based on different error criteria, a normalized mixed norm filtering algorithm that can suppress multi-mode noise is proposed. The optimal filtering algorithm is obtained by adjusting the mixing parameters with the cost function of introducing the mixing parameters. Establish the nonlinear function relationship between the step size and the input SNR. The simulation results show that compared with the Gaussian filter, the robustness of the proposed algorithm is improved significantly in multi-mode noise environment, which can effectively extract the signal.