改进的变误差宽度变阶数LMS算法
发布时间：2019-04-09 18:07
【摘要】：针对自适应滤波器阶数失配问题,提出一种改进的变误差宽度分数阶变阶数LMS算法。该算法中误差宽度函数参数选择不受噪声先验知识的限制,并给出参数选择的依据。对提出的算法分别在高噪声、低噪声以及变化的噪声环境下进行仿真分析,仿真结果表明,该算法在未知系统噪声大小或大小变化的噪声环境能够应用,并且具有良好的性能,尤其在高噪声环境下能够获得较快的阶数收敛速度和较小的稳态误差,因此该算法的应用场合更加广泛。
[Abstract]:To solve the problem of adaptive filter order mismatch, an improved variable error width fractional order variable order LMS algorithm is proposed. In this algorithm, the parameter selection of error width function is not limited by the prior knowledge of noise, and the basis of parameter selection is given. The proposed algorithm is simulated in high noise, low noise and variable noise environment, and the simulation results show that the proposed algorithm can be applied in the noise environment with unknown system noise size or size change, and has good performance, and the simulation results show that the proposed algorithm can be applied in the noise environment with unknown system noise size or change, and the proposed algorithm has good performance. Especially in high noise environment, faster order convergence rate and smaller steady-state error can be obtained, so the application of this algorithm is more extensive.
【作者单位】： 陕西科技大学电气与信息工程学院;陕西科技大学理学院;
【基金】：国家自然科学基金资助项目(61401261) 陕西省教育厅科研项目(11KJ1015)
【分类号】：TN713
，
本文编号：2455411
[Abstract]:To solve the problem of adaptive filter order mismatch, an improved variable error width fractional order variable order LMS algorithm is proposed. In this algorithm, the parameter selection of error width function is not limited by the prior knowledge of noise, and the basis of parameter selection is given. The proposed algorithm is simulated in high noise, low noise and variable noise environment, and the simulation results show that the proposed algorithm can be applied in the noise environment with unknown system noise size or size change, and has good performance, and the simulation results show that the proposed algorithm can be applied in the noise environment with unknown system noise size or change, and the proposed algorithm has good performance. Especially in high noise environment, faster order convergence rate and smaller steady-state error can be obtained, so the application of this algorithm is more extensive.
【作者单位】： 陕西科技大学电气与信息工程学院;陕西科技大学理学院;
【基金】：国家自然科学基金资助项目(61401261) 陕西省教育厅科研项目(11KJ1015)
【分类号】：TN713
，
本文编号：2455411
本文链接：https://www.wllwen.com/kejilunwen/dianzigongchenglunwen/2455411.html
