You are on page 1of 2

# ECE 561: Detection & Estimation Theory

Spring 2013

Homework 1
Assigned: 1/17, Due: 1/24

Problem 1. (Review of conditional probability distributions and expectations). Let U1 , U2 , and V1 be three iid N (0, 1) random variables and W a degenerate random variable equal to (0, 0) with probability 1. Dene the pairs U = (U1 , U2 ) and V = (V1 , 0). Let Y R2 be 1 equal to either U , V , or W , with probability 3 . (a) Which one(s) of the random variables U1 , U2 , U, V1 , V, W, Y has a pdf? a pmf? (b) Give E[ Y (c) Give P [ Y (d) Give E[ Y (e) Give P [ Y
2 ]. 2 2 2

## < 1]. |Y = W ]. < 1 |Y = W ].

Problem 2. (Health Insurance). Find the minimax decision rule for the health insurance problem discussed in class. Problem 3. (Wise investor). Ann has received \$1000 and must decide whether to invest it in the stock market or to safely put it in the bank where it collects no interest. Assume the economy is either in recession or in expansion, and will remain in the same state for the next year. During that period, the stock market will go up 10% if the economy expands, and down 5% otherwise. The economy has a 40% prior probability to be in expansion. An investment guru claims to know the state of the economy and reveals it to Ann. Whatever the state is, there is a probability [0, 1] that the guru is incorrect. Ann will act on the binary information provided by the guru but is well aware of the underlying probabilistic model. Give the Bayesian and minimax decision rules and Anns expected nancial gains in terms of . Perform a sanity check to verify your answer is correct in the extreme cases = 0 and = 1. Problem 4. Find the Bayes rule and the minimum Bayes risk for a binary hypothesis testing problem with equal priors and uniform costs, with p0 (y) = c0 y 2 : |y| 1 0 : else c1 (3 |y|) : |y| 3 0 : else
1

p1 (y) =

## ECE 561 HW1

Spring 2013

where c0 and c1 are constants. Problem 5. A communication system uses binary signaling by sending one of the following two signals: s1 (t) to send one, and s0 (t) to send zero, where s1 (t) = As(t), s0 (t) = As(t), A > 0, and 1 : 0<t<1 s(t) = 0 : else The transmitted signal is corrupted by zero mean white Gaussian noise W (t) with autocorrelation function RW ( ) = 2 ( ), so that the received signal is given by Y (t) = As(t) + W (t) : 1 sent As(t) + W (t) 0 : sent

## The receiver computes the decision statistic

Z=

Y (t)s(t)dt.

The receiver would like to decide whether 0 or 1 was sent, or (if it is not condent about a 0/1 decision), erase the signal. We would like to obtain a Bayesian decision rule for doing this as follows. Let X = {0, 1} and A = {0, 1, e}. The observation is Z. Assume the cost structure 1 a = j, a = 0, 1, 0 a=j C(a, j) = c a=e where j = 0, 1, and where c (0, 1) is the cost of an erasure. (a) Find the Bayes rule for equal priors. Simplify the form of the rule as much as possible and specify its dependence on the problem parameters c, A, and 2 . (b) For d = A = 5, nd c and a corresponding Bayesian decision rule A so that the probability of erasure p1 is twice the probability of error p2 . Compare with the values 1 of p1 and p2 for a Bayesian decision rule B corresponding to c = 2 .