KE36803 Probability and Random Processes

Tutorial 5

Question No. 1
Discrete random variable x is described by the PMF

{
i
K− i=0 ,1 , 2
p X ( i )= 12
0 ot h erwise

(a) Determine the numerical value of K.

(b) Let Cl, and C2 represent 2 successive independent experimental values of random variable x.
Determine that the probability that Cl > C2.
(c) Define R = min(C1, C2) with C1 and C2 as defined in part (b). Determine the expected value of the
random variable R.
(d) How many possible non-negative integer solutions are there to the equation D1 + D2 + D3 ≤ 2?
(e) Let D1, D2 and D3 represent 3 successive independent experimental values of random variable X.
Determine that the probability that D1 + D2 + D3 ≤ 2.
(f) Define R = max(D1, D2) with D1 and D2 as defined in part (e). Determine the variance of the
random variable R.

1
Question No. 2
A communication system has 3 links. Link #2 is connected in parallel with link #3. These two links are
connected in series with link #1. Each link may fail or not fail independently of the other links in
a given interval of time. Define the events
F i={ Component ¿i fails } , i=1 , 2 ,3
Assume that P{Fi} = pi with p1 = .01, p2 = 0.1 and p3 = 0.2.

(a) From all possible individual and mutually exclusive events that make up the sample space, list
out the events that will cause the failure of the communication system.
(b) Determine the probability of failure of the communication system.
(c) Let Mi denote the capacity of link #i in megabits per second, and assume that M 1 = 10, M2 = 6
and M3 = 4. Let the capacity of the system be denoted by X, which in this case is a random
variable due to failures of the links. Derive the probability mass function for the random variable
X.

2
Question No. 3
Messages can be transmitted between Kota Kinabalu and Melaka either directly or via Kuala Lumpur
on communication links whose capacities are given as below:
 capacity from Kota Kinabalu to Kuala Lumpur is 200 MB/s.
 capacity from Kuala Lumpur to Melaka is 40 MB/s.
 capacity from Kota Kinabalu to Melaka is 100 MB/s.

Assume that the three links fail independently with equal probability 1/2. Let X denotes the
communication capacity between Kota KInabalu and Melaka.

(a) Determine the probability that the communication link between Kota Kinabalu and Kuala
Lumpur fails given that the messages failed to be transmitted from Kota Kinabalu to Melaka.
(b) Determine the expected value of X.
(c) Determine the expected value of X given that the messages are transmitted from Kota Kinabalu
to Melaka.

3
Question No. 4
(a) If X is a binomial random variable with parameters N = 4, p = 1/3, what are the mean and
variance of the random variable 2 - 3X?

(b) Select integrated circuits, test them in sequence until you find the first failure and then stop. Let
N be the number of tests. All tests are independent with probability of failure p = 0.1. Consider
the condition B = {N  20). Determine E[N|B].

(c) A fair coin is tossed repeatedly until a Head occurs. N denotes the number of tosses. Given that
the event P(N >5) occurred, what is the expected value of N?

4
Question No. 5
Suppose that a cellular phone costs RM20 per month with 30 minutes of use included and that each
additional minute of use costs RM0.50.

(a) If the number of minutes you use each month is a geometric random variable M with expected
value of E[M] = 30 minutes, determine the PMF of C, the cost of the phone for one month.

(b) Determine the expected monthly cost E[C].

(c) A new cellular phone billing costs RM15 per month plus RM1 for each minute of use. If the
number of minutes you use each month is a geometric random variable with mean 1/p. What
values of p is this billing plan preferable to the original billing plan?

5
Question No. 6
Over the years that an absent-minded Dr. Siva has been traveling, he has taken mid-semester breaks
in either Kuching or Ipoh. When he went on mid-semester break this year, he could not
remember where he was, but he knew that it must be either Kuching or Ipoh. Being a part-time
Feng Shui Master, he knew the following likelihoods of the weather in Kuching and Ipoh during
the mid-sem break:
Sunny Cloudy Rainy
Kuching 0.50 0.20 0.30
Ipoh 0.20 0.30 0.50

Referring to his calendar, Dr. Siva estimated that the prior probability of his being in Kuching on any
given mid-sem break is 0.70. He thus concludes that the prior probability that he is in Ipoh is
0.30. Dr. Siva would like to use this information to make a decision about his current location.
(a) How many possible decision rules exist for this problem?
(b) Determine the MAP decision rule.
(c) Determine the probability of error associated with the maximum likelihood decision rule.
(d) Determine the probability that Dr. Siva is in Kuching given that it is sunny day.
(e) Are the location of Dr. Siva's visit and the weather independent of each other? Justify your
answer.

6
Question No. 7
There are three buckets: one grey, one white, and one black. The grey bucket contains 5 white and 3
black balls. The white bucket contains 6 white and 4 red balls. The black bucket contains 4 black and
6 red balls. A ball is first drawn from the grey bucket. The color of that ball determines which bucket
a second ball is drawn from; the second ball is drawn from the bucket with the same color as the
first ball. For example, if the first ball is white, then the second ball is drawn from the white bucket.
Let X= 0 if the second ball is white, X= 1 if the second ball is red, X= 2 if the second ball is black.

(a) Derive the probability mass function (pmf) of X.

(b) Let H0 be the hypothesis the first ball is white and H1 be the hypothesis the first ball is black.
(i) Obtain the maximum likelihood rule for selecting a hypothesis given X.
(ii) Obtain the MAP rule for selecting a hypothesis given X.

7
Question No. 8
Consider a binary hypothesis testing problem where under H 1 the random variable X is Geometric
with expected value of 2, while under H 0, it is instead uniform (its pmf is constant) on the
set of positive consecutive integers {2, 3, 4, 5}.
(a) Obtain the Maximum Likelihood rule. To be definite, ties are broken in favor of H1.

(b) Under the Maximum Likelihood rule from part (a), obtain the probability of missed detection.

(c) Let P(H0) = p. For what value of p will H 0 be always be chosen under MAP rule on the set of
positive consecutive integers {2, 3, 4, 5}?