欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
针对α稳定分布噪声环境下的时延估计问题,对最大似然加权估计法进行改进,给出了三种高效实用的新算法。首先,以分数低阶统计量为基础,提出了一种基于分数低阶统计量的最大似然时延估计算法（FLO-ML算法）;其次,通过函数变换,提出了两种不依赖于分数低阶统计量的新算法（Log-ML算法和UDE-ML算法）;进一步,本文还详细讨论了三种新算法的适用范围及计算复杂度。仿真分析表明,三种新算法均能在分数低阶α稳定分布噪声环境下实现准确的时延估计,其性能优于同类算法,同时三种新算法都能在传统高斯噪声环境下保持良好的稳健性。
英文摘要：
In regards of the time delay estimation problem under α stable noise environment, the weighted esti- mation method of maximum likelihood is improved, and put forward three new efficient and practical algorithms. First, a FLO-ML algorithm is proposed based on fractional lower order statistics; second, by function transforma- tion, two new algorithms are put forward based on functional transformation, which are Log-ML algorithm and UDE- ML algorithm, they do not depend on fractional lower order statistics; third, the paper also discusses the scope and complexity of the three new algorithms in detail. Simulation results show that the three new algorithms can achieve accurate delay estimation in fractional lower order of αstable noise environment, and their performance is better than other algorithms, at the same time, the three new algorithms can maintain a good robustness under the traditional Gaussian noise environment.
同期刊论文项目
同项目期刊论文
期刊信息
位置：