摘要
提出了用对数正态分布描述噪声相关性偏态分布的最佳模型。首先通过理论分析和实验发现,在(0,1]区间β分布、Γ分布和对数正态分布可用于描述噪声相关性的偏态分布;然后分别利用这3种分布的概率密度函数(PDF)模拟实际的噪声相关性偏态分布曲线,曲线的形态与实际的噪声相关性偏态分布形态的相似程度的比较表明,采用对数正态分布描述噪声相关性偏态分布的效果最佳;最后利用最大熵方法证明了模型的最佳性。实验结果表明,本文提出的模型与另外两种模型相比,可使最小错误率降低70%以上。
The optimal model for the asymmetrical distribution of noise correlation is proposed using the logarithmic normal distribution.It is found that beta distribution,gamma distribution and logarithmic normal distribution can be used to describe the asymmetrical distribution of noise correlation within（0,1] by the theoretical analysis and experiments,and then curves of the actual asymmetrical distribution of noise correlation are simulated by using the three probability density functions respectively.It is explained that the best results can be achieved using the logarithmic normal distribution according to the comparison between the simulated and actual curves of asymmetrical distribution of noise correlation,and it is proved by the maximum entropy method.Experimental results show that this model can reduce the least false rate over 70% compared with the other models.
出处
《光电子．激光》
EI
CAS
CSCD
北大核心
2010年第12期1877-1880,共4页
Journal of Optoelectronics·Laser
基金
福建省教育厅科技基金资助项目(JA08245)
关键词
噪声相关性
偏态分布
概率密度函数(PDF)
最大熵方法
noise correlation
asymmetrical distribution
probability density function
maximum entropy method
