测量数据处理中的Bayes理论与最大熵方法

Bayes Theorem and Maximum Entropy Principle in the Measurement Data Processing

摘要: 以Bayes理论和Jaynes的最大熵方法为基础,分析了测量数据概率空间的特点,提出了将Bayes理论与最大熵方法结合,利用最大熵方法求取Bayes理论中的先验分布和似然函数的测量数据PDF反演方法。实验结果证明此方法的反演结果优于单独运用Bayes或最大熵方法计算的反演结果。

Abstract: On the base of the Bayes theorem and Maximum entropy principle, the paper analyzes the measurement data characteristics and designs the inversion method of measurement data PDF by the combination of the Bayes theory and Maximum principle. The prior distribution and the likelihood function of measurement data PDF in the Bayes theorem are calculated using the Maximum entropy principle. The results are proved by the experiment of measuring frequency.