数字信号处理chapter3.ppt

Statistical and Adaptive Signal ProcessingChapter 3;Chapter 3Random Variables, Vectors, and Sequences;3.1 Random Variables;3.1 Random Variables;3.1 Random Variables;3.1 Random Variables;Cumulative Distribution Function (cdf) Fx(x) The probability set function Pr{x(ζ) ≤ x} is a function of the set {x(ζ) ≤ x}, but it is also a number that varies with x. Hence it is also a function of a point x on the real line R. This point function is the well-known cumulative distribution function (cdf) Fx(x) of a random variable x(ζ) and is defined by Probability Density Function (pdf) fx(x) Note that the pdf fx(x) is not the probability, but must be multiplied by a certain interval Δx to obtain a probability, that is, ;Probability Density Function (pdf) fx(x) Integrating both sides of (3.1.2), we obtain Probability Mass Function (pmf) pk For discrete-valued random variables, we use the probability mass function (pmf) pk, defined as the probability that random variable x(ζ) takes a value equal to xk, or Properties and Applications; In practice, it is desirable to summarize some of the key aspects of a density function by using a few numbers rather than to specify the entire density function. These numbers, which are called statistical averages or moments, are evaluated by using the mathematical expectation operation. Mathematical expectation We will follow the engineering practice of using the expression for the continuous random variable. The mean ?x can be regarded as the “location” (or the “center of gravity”) of the density fx(x).;3.1.2 Statistical Average; One important property of expectation is that it is a linear operation, Let y(ζ)= g[x(ζ)] be a random variable obtained by transforming x(ζ) through a suitable function. Then the expectation of y(ζ) is given by ;3.1.2 Statistical Average;3.1.2 Statistical Average;3.1.2 Statistical Average;3.1.2 Statistical Average;