刘诚,邱天爽,李景春,李蓉.脉冲噪声环境下基于最大相关熵准则的仿射投影算法[J].,2018,58(1):86-91 |
脉冲噪声环境下基于最大相关熵准则的仿射投影算法 |
Affine project algorithm based on maximum correntropy criterion for impulsive noise environment |
|
DOI:10.7511/dllgxb201801013 |
中文关键词:脉冲噪声最大相关熵准则仿射投影算法自适应滤波 |
英文关键词:impulsive noisemaximum correntropy criterionaffine project algorithmadaptive filtering |
基金项目:国家自然科学基金资助项目(61671105,61172108,61139001). |
|
摘要点击次数:1112 |
全文下载次数:1218 |
中文摘要: |
在许多工程技术应用中,噪声常呈现较强的冲激性,在数学上可用稳定分布模型来表示.常规的自适应滤波算法对于稳定分布噪声的鲁棒性较差.最大相关熵准则的提出,可以有效地改善在稳定分布噪声条件下自适应滤波算法的性能.以对称α稳定分布噪声为脉冲噪声模型,使用最大相关熵准则作为代价函数,提出基于最大相关熵准则的仿射投影算法,将前一次迭代的计算误差作为自变量代入相关熵函数中,使得改进后的仿射投影算法在脉冲噪声环境下取得良好的收敛效果,并通过实验证明了新算法在强脉冲噪声以及广义信噪比较低的情况下有较快的收敛速度和较好的收敛性能. |
英文摘要: |
In many engineering applications, noise often shows strong impulsive, and is represented by a stable distribution model in mathematics. Traditional adaptive filtering algorithm is not robust in stable distribution noise. Maximum correntropy criterion can effectively improve the performance of the adaptive filtering algorithm under impulsive noise conditions. Applying symmetric α stable noise as the impulsive noise model, maximum correntropy criterion as the cost function, an affine project algorithm based on maximum correntropy criterion is proposed. By using a prior error as independent variable in the correntropy function, the improved affine project algorithm obtains a good convergence effect in impulsive noise environment. Experimental results show that the new algorithm has good convergence speed and performance under strong impulsive noise and low generalized signal-to-noise ratio conditions. |
查看全文查看/发表评论下载PDF阅读器 |
| --> 关闭 |