Академический Документы
Профессиональный Документы
Культура Документы
The following homework set has two types of problems. Those labeled SS are self-study
problems that do not need to be turned in. However, there are problems that I suggest you complete
to ensure that you understand all the material in the class. The problems not labeled SS should be
submitted for a grade. In general, the SS problems are easier than the graded problems, so they
can be used to practice and build expertise before doing the graded problems.
SS-1. You are sent as part of a UN inspection team to investigate whether North Korea has weapons-grade
plutonium (WGP) in a reactor. Let W denote the event that WGP is present.
You take measurements with a Geiger counter and decide that WGP is detected if some
threshold is exceeded. Suppose that the threshold is set so that the probability of correct
detection, P(D|W ) is 0.9, and the probability of false detection, P(D|W ) is 0.25.
Suppose that the probability that WGP is present is 0.3.
It may help to draw a diagram showing the transition probabilities.
(a) What is the probability that WGP is detected? (Anwer: 0.445) Not detected? (Answer:
0.555)
(b) If you detect WGP, what is the probability that it is actually present? (Answer: 0.607)
(c) If you detect WGP, what is the probability that it is not present? (The conditional
probability of false alarm). (Answer: 0.393)
(d) If you do not detect WGP, what is the probability that it was actually there? (The
conditional probability of a miss). (Answer: 0.054)
(e) What is the overall probability that you make the wrong conclusion based on your
measurements? (Answer: 0.205)
1. Consider the binary communication system shown in Figure 1.
A0
p00
B0
p10
p01
A1
p11
B1
(b) Find all the possible MAP decision rules when p01 = 1/8 and p10 = 1/6. Specify the
range of p0 for which each MAP decision rule holds.
2. From Random Signal Analysis in Engineering Systems by John J. Komo (slightly modified)
For the digital communication system shown in Figure 2, where P(A0) = 0.6 and P(A1) =
0.4, completely specify the MAP decision rule and calculate the overall probability of error
under that rule.
A0
0.5
B0
0.25
0.1
0.25
B1
0.3
A1
0.6
B2
(d) Using the assumptions of the previous part of this problem, what is the probability of
having at least one week with three or more shark attacks during a given year?
4. (Modified from Stark and Woods, 3rd ed.)
(Independence of events in disjoint arrivals for Poisson law) The average number of cars
arriving at a tollbooth per minute is and the probability of k cars in the interval (0, T ) is
T
P(k; 0, T ) = e
[ T ]k
.
k!
Consider two disjoint, that is nonoverlapping intervals, say (0,t1 ) and (t1 , T ). Then for the
Poisson law:
P[n1 cars in (0,t1 ) and n2 cars in (t1 , T )] = P[n1 cars in (0,t1 )]P[n2 cars in (t1 , T )]
Thus events in disjoint intervals are independent.
(a) Show that P[n1 cars in (0,t1 )|n1 + n2 cars in (0, T )] is not a function of .
(b) In (a) let T = 2, t1 = 1 and n1 = 5, n2 = 5. Compute P[5 cars in (0, 1)|10 cars in (0, 2)]
(c) Show that the probability that you found in part (a) is a binomial probability. In other
words, show that the probability can be written as
N n1
p (1 p)Nn1 .
n1
Express N and p as functions of n1 , n2 ,t1 and t2 .
(d) Give an intuitive explanation for the formula for p in part (c).
5. Suppose we choose a number at random from the interval S = [0, 5], and denote this value
by X.
(a) Let Y be the largest integer less than or equal to X.
i. Consider the event A = {Y 3}. Find the equivalent event, A0 , in S and the
probability of A.
ii. Consider the event B = {Y = 3}. Find the equivalent event, B0 , in S and the
probability of B.
(b) Let Z = X + 1.
i. Consider the event C = {Z 3}. Find the equivalent event, C0 , in S and the
probability of C.
ii. Consider the event D = {Z = 3}. Find the equivalent event, D0 , in S and the
probability of D.