不規則騒音のレベル変動に関する統計的研究(IV) : 一次相関情報のある実用的検出方法

概要

論文の詳細を見る
First, two general expressions of the conditional probability density function in the form of expansion series are calculated by referring to the definition of conditional probability from two expansion expressions of the bivariate joint probablity density function derived previously in the form of orthonormal expansion series such as the statistical Hermite expansion or the statistical Laguerre expansion. However, for the purpose of getting information on correlation of the first order, we had better find the expression of conditional average instead of using the above conditional probability density functions. Thus, two explicit expressions of the correlation function of the first order are prmented in the form of expansion series from the equations of conditional average. We must give attention to the fact that a universal form comes first in these expansions and factors determining the form of the marginal one-variate probability density function and the correlation functions of high orders are reflected in the compensation terms. Further, a simple expression of the correlation of the first order related to the conditional average can be approximately derived in a single form as a practical method for the detection of information on correlation of the first order and is also experimentally verified.
社団法人日本音響学会の論文
1968-05-30