Question on determining the posterior pdf

40 Views Asked by At

Can someone tell me how the pdf of noise (w) is equivalent to the conditional pdf of observations (x) given A, assuming noise is independent of A for the equation x[n]=A+w[n] where A is the mean (and a random variable)?

This is done in order to determine the posterior pdf using Bayes' rule.From the book "Fundamentals of Estimation and Detection by Stephen Kay

1

There are 1 best solutions below

1
On

Weird notations...

One has the following fact. If $W$ and $A$ are independent random variables, and if $X = h(W, A)$ for some measurable function $h$, then the conditional distribution of $X$ given $A = a$ is the distribution of the random variable $h(W, a)$ (I should talk about regular conditional distribution to be rigorous). Here $h(w, a) = w + a$. Now it is easy to see that if $f_W$ is the pdf of $W$, then the random variable $Y := W + a$ has pdf $f_Y$ given by $f_Y(y) = f_W(y - a)$.