Solutions to Problem Set 9 in ECE 313 at University of Illinois, Spring 2008, Assignments of Statistics

The solutions to problem set 9 in the electrical and computer engineering (ece) 313 course at the university of illinois, spring 2008. It covers various probability distributions and statistical concepts, including poisson random variables, exponential random variables, and normal distributions. Students can use this document to check their understanding of the concepts covered in the problem set.

Typology: Assignments

Pre 2010

Uploaded on 03/16/2009

koofers-user-m96
koofers-user-m96 🇺🇸

10 documents

1 / 2

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
University of Illinois Spring 2008
ECE 313: Solutions to Problem Set 9
1. (a) E[Y] = E[X2] = Z+1
1
u21
2du =1
3.E[Y2] = E[X4] = Z+1
1
u41
2du =1
5.
Hence, var(Y) = E[Y2](E[Y])2=1
51
9=4
45.
(b) E[Z] = E[g(X)] = Z0
1u21
2du +Z+1
0
u21
2du =1
6+1
6= 0.
(c) Utakes on integer values 1. V= sin(πU/2) = 0 if Uis even. On the other hand, if
U= 4k+ 1, then V= +1 while if U= 4k1, then V=1. We get
E[V] =
X
k=1
sin(πk/2) 1
2k
=1
21
8+1
32 · ·· =1
21x+x2 · ·· =1
2(1 + x)
where x=1
4E[V] = 2
5.
2. (a) E[R] = Z1
0
u·2u du =2
3.E[V] = Z1
0
4π
3u3·2u du =8π
15 .E[A] = Z1
0
4πu2·2u du = 2π.
The average sphere does not have average volume or average surface area. More generally,
E[g(X)] 6=g(E[X]).
(b) For 0 <u<1, un> un+1.
Hence, Z1
0
un·fR(u)du =E[Rn]>Z1
0
un+1 ·fR(u)du =E[Rn+1].
3. I’m considering, I’m considering!
(a) The number of arrivals in (0,4] is N(0,4], a Poisson random variable with parameter
4λ. Hence, E[N(0,4]] = 4λ.
(b) The event {N(0,3] = 3}∩{N(2,6] = 0}is the same as {N(0,2] = 3}∩{N(2,6] = 0}
and thus P[{N(0,3] = 3}∩{N(2,6] = 0}] = P[{N(0,2] = 3}∩{N(2,6] = 0}]. We get
P[{N(0,3] = 3}∩{N(2,6] = 0}] = (2λ)3
3! exp(2λ)×exp(4λ) = 4λ3
3exp(6λ)
by independence of Poisson variables on disjoint intervals of time.
(c) The number of arrivals in (0,6] is N(0,6], a Poisson random variable with parameter 6λ.
If it is observed that N(0,6] = k, then the maximum-likelihood estimate of the Poisson
parameter 6λis k(cf. Problem 2(c) of Problem Set 5), and hence the maximum-
likelihood estimate of the arrival rate is k/6 = # of arrivals/length of interval which is
exactly what one would expect.
(d) P{N(0, t)1}= 1 P{N(0, t) = 0}= 1 exp(λt)=12twhen λ= ln 2.
4. (a) The inter-arrival time in a Poisson process with arrival rate λ(time between two suc-
cessive chalk breaks on this instance) is an exponential random variable with parameter
λ. Hence, the average length of time between successive chalk-breaks is the mean of this
exponential random variable, which is λ1= 10 minutes.
(b) The number of times that Todd breaks the chalk during a 50 minute lecture is a Poisson
random variable N(0,50] with parameter λ×50 = 5 and mean value E[N(0,50]] = 5.
pf2

Partial preview of the text

Download Solutions to Problem Set 9 in ECE 313 at University of Illinois, Spring 2008 and more Assignments Statistics in PDF only on Docsity!

University of Illinois Spring 2008

ECE 313: Solutions to Problem Set 9

  1. (a) E[Y] = E[X 2 ] =

− 1

u^2

du =

3.^ E[Y

2 ] = E[X 4 ] =

− 1

u^4

du =

Hence, var(Y) = E[Y^2 ] − (E[Y])^2 =

5 −^

9 =^

(b) E[Z] = E[g(X )] =

− 1

−u^2

du +

0

u^2

du = − 1 6

+^1

(c) U takes on integer values ≥ 1. V = sin(πU/2) = 0 if U is even. On the other hand, if U = 4k + 1, then V = +1 while if U = 4k − 1, then V = −1. We get E[V] =

∑^ ∞

k=

sin(πk/2)

)k =^1 2

+^1

− · · · =^1

[

1 − x + x^2 − · · ·

]

2(1 + x) where x =^14 ⇒ E[V] =^25.

  1. (a) E[R] =

0

u · 2 u du =^2 3

. E[V] =

0

4 π 3

u^3 · 2 u du =^8 π 15

. E[A] =

0

4 πu^2 · 2 u du = 2π. The average sphere does not have average volume or average surface area. More generally, E[g(X )] 6 = g(E[X ]). (b) For 0 < u < 1, un^ > un+1. Hence,

0

un^ · fR(u) du = E[Rn] >

0

un+1^ · fR(u) du = E[Rn+1].

  1. I’m considering, I’m considering! (a) The number of arrivals in (0, 4] is N (0, 4], a Poisson random variable with parameter 4 λ. Hence, E[N (0, 4]] = 4λ. (b) The event {N (0, 3] = 3} ∩ {N (2, 6] = 0} is the same as {N (0, 2] = 3} ∩ {N (2, 6] = 0} and thus P [{N (0, 3] = 3} ∩ {N (2, 6] = 0}] = P [{N (0, 2] = 3} ∩ {N (2, 6] = 0}]. We get

P [{N (0, 3] = 3} ∩ {N (2, 6] = 0}] = (2λ)

3 3!

exp(− 2 λ) × exp(− 4 λ) =^4 λ

3 3

exp(− 6 λ)

by independence of Poisson variables on disjoint intervals of time. (c) The number of arrivals in (0, 6] is N (0, 6], a Poisson random variable with parameter 6λ. If it is observed that N (0, 6] = k, then the maximum-likelihood estimate of the Poisson parameter 6λ is k (cf. Problem 2(c) of Problem Set 5), and hence the maximum- likelihood estimate of the arrival rate is k/6 = # of arrivals/length of interval which is exactly what one would expect. (d) P {N (0, t) ≥ 1 } = 1 − P {N (0, t) = 0} = 1 − exp(−λt) = 1 − 2 −t^ when λ = ln 2.

  1. (a) The inter-arrival time in a Poisson process with arrival rate λ (time between two suc- cessive chalk breaks on this instance) is an exponential random variable with parameter λ. Hence, the average length of time between successive chalk-breaks is the mean of this exponential random variable, which is λ−^1 = 10 minutes. (b) The number of times that Todd breaks the chalk during a 50 minute lecture is a Poisson random variable N (0, 50] with parameter λ × 50 = 5 and mean value E [N (0, 50]] = 5.

(c) From part (b), we get that P {N (0, 50] = 6} =^5

6 6!

exp(−5). Now, for 0 ≤ k ≤ 6,

P {{N (0, 25] = k} | {N (0, 50] = 6}} =

P {{N (0, 25] = k} ∩ {N (0, 50] = 6}} P {N (0, 50] = 6}} = P^ {{N^ (0,^ 25] =^ k} ∩ {N^ (25,^ 50] = 6^ −^ k}} P {N (0, 50] = 6}} = P^ {N^ (0,^ 25] =^ k}P^ {N^ (25,^ 50] = 6^ −^ k}} P {N (0, 50] = 6}}

=

(2.5)k k! exp(−^2 .5)^ ×^

(2.5)^6 −k (6−k)! exp(−^2 .5) 56 6! exp(−5) =

k

)k ( 1 2

) 6 −k

Thus, the conditional pmf of N (0, 25] given that {N (0, 50] = 6} is a binomial pmf with parameters (6, 12 ) and hence the expected value is 3.

  1. (a) P {X < 0 } = Φ

= Φ(5) = 1 − Q(5).

(b) P {− 10 < X < 5 } = Φ

2

2

1 2 −^ Q(7.5). (c) P { |X | ≥ 5 } = P { X ≤ − 5 } + P { X ≥ 5 } = Φ

2

2

1 − Φ(7.5) = 1 − Q(2.5) + Q(7.5).

(d) P {X 2 − 3 X +2 > 0 } = P {(X −1)(X −2) > 0 } = P {X < 1 }+P {X > 2 } = Φ

2

2

= Φ(5.5) + 1 − Φ(6) = 1 − Q(5.5) + Q(6).

  1. Let X ∼ N (0. 9 , 0. 0032 ) denote the width (in microns) of the trace.

(a) {X < 0. 9 − 0. 005 } or {X > 0 .9 + 0. 005 } for a trace to be deemed defective. P { |X − 0. 9 | > 0. 005 } = 2Φ(− 0. 005 / 0 .003) = 2Φ(− 1. 666.. .) = 2Q(1. 666.. .) ≈ 0 .095. (b) We need to find the maximum value of σ such that 2Q(0. 005 /σ) ≤ 10 −^2. Since Q(2.575) ≈ 0 .005, we get that 0. 0005 /σ > 2. 575 , that is, σ ≤ 0. 005 / 2. 575 ≈ 0 .00194.

  1. Y (f ) = H(f )X(f ) = rect(f /2) exp(−πf 2 ), and y(0) =

−∞

Y (f ) df =

− 1

exp(−πf 2 ) df. But, the integrand is the Gaussian pdf N (0, (2π)−^1 ). Hence, y(0) = Φ(

2 π) − Φ(−

2 π) = 2Φ(

2 π) − 1 ≈ 0. 9826