论文部分内容阅读
为解决脉冲噪声下最小均方误差自适应时间延迟估计算法估计性能的退化问题,以对称α稳定分布模型描述脉冲噪声,提出最小均方Sigmoid误差自适应时间延迟估计算法.该算法通过对误差信号求取Sigmoid变换,抑制了较大误差对估计结果的影响.以最小均方Sigmoid误差代替最小均方误差作为优化准则,迭代模拟信道延迟效应的滤波器权系数,其收敛时峰值的位置就是所要估计的时间延迟.仿真结果验证了该算法在高斯和非高斯对称α稳定分布噪声条件下的优良估计性能,说明最小均方Sigmoid误差是一种韧性的最优准则.
In order to solve the problem of degradation of estimation performance of the minimum mean square error adaptive time delay estimation algorithm under impulsive noise, impulse noise is described by a symmetrical α-stable distribution model and a minimum mean square Sigmoid error adaptive time delay estimation algorithm is proposed. The Sigmoid transform is used to suppress the influence of large errors on the estimation results.With the smallest mean square Sigmoid error instead of the least mean square error as the optimization criterion, the filter weight coefficient of the channel delay effect is iterated, and the peak position of the filter is the desired And the estimated time delay.The simulation results verify the good performance of the proposed algorithm in the Gaussian and non-Gaussian symmetric αstable distribution noise conditions, indicating that the minimum mean square Sigmoid error is a toughness optimal criterion.