Digital Information Age An Introduction to

Electrical Engineering 2nd Edition Roman

Full download at link: https://testbankpack.com/p/solution-manual-for-digital-
information-age-an-introduction-to-electrical-engineering-2nd-edition-by-
roman-kuc-isbn-1305077717-9781305077713/

Chapter 6

Modeling Random Data & Noise

6.1 Problems
Problem 6.1 (Sample mean, variance, and SD) Let the RV X produce the following nx = 10 real-
izations:
1.7, −0.8, −0.6, 0.5, 0.3, −0.5, 0.8, −0.1, 0.9, −0.7
Compute the sample mean, variance, and SD.
(ans:
Sample mean:
1.7 − 0.8 − 0.6 + 0.5 + 0.3 − 0.5 + 0.8 − 0.1 + 0.9 − 0.7
Ave(Xi ) =
= 0.15
10
Sample variance: V ar(Xi ) = Ave(X i2 ) − [Ave(Xi )]2
2 2.89 + 0.64 + 0.36 + 0.25 + 0.9 + 0.25 + 0.64 + 0.01 + 0.81 +
Ave(X
0.49 )=
i = 0.724
10
V ar(Xi ) = 0.724 − 0.0225 = 0.7015
Sample SD: √
SD(Xi ) = V ar(Xi ) = 0.838
)

Problem 6.2 (Probability that Y lies in [0.25,0.75)) What is the probability that the RV Y that is uni-
formly distributed over [0,1), will lie in the range [0.25, 0.75)?

(ans: The area from x = 0.25 to x = 0.75 = 0.5.

)

© 2015 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 Problem 6.3 (Probability that Y lies in [0.25,2)) What is the probability that the RV Y that is uni-
formly distributed over [0,1), will lie in the range [0.25,2)?

(ans: The area from x = 0.25 to x = 2 under the PDF that extends over [0,1] is the same as the area
from x = 0.25 to x = 1 = 0.75 .
)

55

© 2015 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 56 CHAPTER 6. MODELING RANDOM DATA & NOISE

Problem 6.4 (Simulating random die toss) Starting with realizations of Y , how would you form T ,
the RV that simulates the result of a fair die toss, which equals the number of dots showing on the top
face? Sketch the PMF of T .

(ans:
Y extends over [0,1).
6Y extends over [0,6).
T = Ceil(6Y ) is the integer set [1,6].
The PMF of T :

1 2 3 4 5 6
)

Problem 6.5 (Adding 1 to Y ) Starting with RV Y , we add 1 to each value we observe to form
X = 1+ Y

Compute µX , σX
2 and sketch p (x).
X

(ans:
=0.5
z}|{
µX = µY +1 = µY +1 = 1.5
1
σ2 2 2
X = σY +1 = σY =
12
pX (x):

0 1 2

Problem 6.6 (Adding 1 to G) Starting with standardized RV G, add 1 to each value to form
X = 1+ G

Compute µX , σX
2
and sketch pX (x).

(ans:
=0
z}|{
µX = µG+1 = µG +1 = 1

© 2015 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 6.1. PROBLEMS 57

σ2X = σ2G+1 = σG2 = 1

pX (x):

σ =1

-1 0 1 2 3

Problem 6.7 (Multiplying Y by 2) Starting with RV Y that is uniformly distributed over [0,1), multi-
ply each value to form
Z = 2Y

Compute µZ , σ2Z and sketch pZ (x).

(ans:
=0.5
z}|{
µZ = µ2Y = 2 µY = 1

2 4 1
σ2 2
Z = σ2Y = 4σY = =
12 3
pZ (x):

1/2

0 1 2

Problem 6.8 (Multiplying G by 2) Starting with standardized Gaussian RV G, multiply each value
by 2 to form
Z = 2G

Compute µZ , σ2Z and sketch pZ (x).

(ans:
=0
z}|{
µZ = µ2G = 2 µG = 0

σ2Z = σ2G
2 = 4σ 2 = 4
G

© 2015 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 58 CHAPTER 6. MODELING RANDOM DATA & NOISE

p (x)
G σ =1

-2 -1 0 1 2

p (x)
Z

σ =2

-4 -3 -2 -1 0 1 2 3 4

Problem 6.9 (Summing two independent standardized Gaussian RV s) Starting with independent stan-
dardized Gaussian RV s G1 and G2 , form

V = 2(G1 + 1) − (G2 − 1)
Compute µV , σV2 and pV (x).

(ans:
V = 2(G1 + 1) − (G2 − 1) = 2G1 + 2 − G2 + 1 = 2G1 − G2 + 3
=0 =0
z}|{ z}|{
µV = µ2G1 −G2 +3 = µV = 2 µG − µG +3 = 3
1 2

=1 =1
z}|{ z}|{
2 2 2 2
σ2 2

V = σ2G1 −G2 +3 = σ2G1 −G2 = 4 σG1 +(−1) σG2 = 5

When adding Gaussian RV , the resulting P DF remains Gaussian and is completely specified by the
mean and standard deviation.

p (x)
G σ =1

-2 -1 0 1 2
p (x)
V
σ =(5)1/2 = 2.23

-1 0 1 2 3 4 5 6 7

© 2015 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 6.2. EXCEL PROJECTS 59

6.2 Excel Projects

Project 6.1 (Random numbers using RAND()) Extend Example 13.28 to generate nY = 25 random
numbers Yi for 0 ≤ i ≤ 24, and compute the sample averages. Observe how the sample averages
change as a new set of random numbers is generated.

(ans:

A B C D
1 i Yi
2 0 0.683
3 1 0.875 Sample Ave=
4 2 0.279 0.487
5 3 0.380
6 4 0.606 Sample SD=
7 5 0.347 0.306
8 6 0.716
9 7 0.072 Sample Var=
10 8 0.397 0.094
11 9 0.876
12 10 0.216
13 11 0.785
14 12 0.985
15 13 0.470
16 14 0.360
17 15 0.245
18 16 0.029
19 17 0.772
20 18 0.014
21 19 0.232
22 20 0.833
23 21 0.593
24 22 0.099
25 23 0.955
26 24 0.357
A B C D
1 i Yi
2 0 =RAND()
3 =A2+1 =RAND() Sample Ave=
4 =A3+1 =RAND() =AVERAGE(B2:B26)
5 =A4+1 =RAND()
6 =A5+1 =RAND() Sample SD=
7 =A6+1 =RAND() =STDEV.S(B2:B26)
8 =A7+1 =RAND()
9 =A8+1 =RAND() Sample Var=
10 =A9+1 =RAND() =VAR(B2:B26)

© 2015 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 60 CHAPTER 6. MODELING RANDOM DATA & NOISE

Project 6.2 (Two Random Dice) Extend Example 13.30 to simulate the toss of two dice, D1i and D2i
for 0 ≤ i ≤ 24, and add the observed values together. Observe how the sums change as another set of
random numbers is generated.

(ans:
A B C D E F
1 i Y1i Y2i D1i D2i Sum
2 0 0.424 0.787 3 5 8
3 1 0.809 0.045 5 1 6
4 2 0.324 0.692 2 5 7
5 3 0.801 0.422 5 3 8
6 4 0.011 0.703 1 5 6
7 5 0.349 0.376 3 3 6
8 6 0.073 0.294 1 2 3
9 7 0.965 0.536 6 4 10
10 8 0.573 0.028 4 1 5
11 9 0.606 0.720 4 5 9
12 10 0.812 0.900 5 6 11
13 11 0.719 0.280 5 2 7
14 12 0.554 0.141 4 1 5
15 13 0.210 0.955 2 6 8
16 14 0.655 0.293 4 2 6
17 15 0.121 0.438 1 3 4
18 16 0.791 0.885 5 6 11
19 17 0.701 0.089 5 1 6
20 18 0.865 0.534 6 4 10
21 19 0.593 0.245 4 2 6
22 20 0.757 0.224 5 2 7
23 21 0.484 0.564 3 4 7
24 22 0.796 0.686 5 5 10
25 23 0.083 0.004 1 1 2
26 24 0.216 0.561 2 4 6
A B C D E F
1 i Y1i Y2i D1i D2i Sum
2 0 =RAND() =RAND() =CEILING(6*B2,1) =CEILING(6*C2,1) =D2+E2
3 1 =RAND() =RAND() =CEILING(6*B3,1) =CEILING(6*C3,1) =D3+E3
4 2 =RAND() =RAND() =CEILING(6*B4,1) =CEILING(6*C4,1) =D4+E4
5 3 =RAND() =RAND() =CEILING(6*B5,1) =CEILING(6*C5,1) =D5+E5
6 4 =RAND() =RAND() =CEILING(6*B6,1) =CEILING(6*C6,1) =D6+E6
7 5 =RAND() =RAND() =CEILING(6*B7,1) =CEILING(6*C7,1) =D7+E7
8 6 =RAND() =RAND() =CEILING(6*B8,1) =CEILING(6*C8,1) =D8+E8
9 7 =RAND() =RAND() =CEILING(6*B9,1) =CEILING(6*C9,1) =D9+E9
10 8 =RAND() =RAND() =CEILING(6*B10,1) =CEILING(6*C10,1) =D10+E10

© 2015 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 6.2. EXCEL PROJECTS 61

Project 6.3 (Histogram of uniformly-distributed random numbers) Using Example 13.32 as a guide
generate a histogram of ten thousand random numbers produced by RAND().

(ans: The problem involves setting up the proper bin intervals in C and computing the bin number,
corresponding to the row number in D, from the random number generated in A2. The example shows
dividing the interval [0,1) into 10 bins, and including the adjacent bins to illustrate their zero counts.
Note that FLOOR must be used, as ROUND produces half-counts in the two limit bins.

A B C D A B C D
1 Yi= bin count 1 Yi= bin count
2 0.53448142 0.1 0 2 = RAND() 0.1 0
3 0 929 3 =C2+0.1 1000
4 i= 0.1 1058 4 i= =C3+0.1 1006
5 10000 0.2 1025 5 10000 =C4+0.1 1000
6 0.3 972 6 =C5+0.1 1010
7 nX= 0.4 944 7 nX= =C6+0.1 952
8 10000 0.5 1028 8 10000 =C7+0.1 989
9 0.6 1011 9 =C8+0.1 975
10 bin#= 0.7 963 10 bin#= =C9+0.1 1009
11 8 0.8 1032 11 =3+FLOOR(10*A2,1) =C10+0.1 1046
12 0.9 1038 12 =C11+0.1 1013
13 Hist 1 0 13 Hist =C12+0.1 0
14 1.1 0 14 =C13+0.1 0

Sub hist()
Dim Val As Integer ’ define an integer variable
Range("D2:D14").Value = 0 ’ reset bin counts
Range("A5").Value = 0 ’ reset counter i
Do While Range("A5").Value < Range("A8").Value ’ loop for nX times
Val = Range("A11").Value ’ get bin value
If Val > 1 Then ’ if valid bin value
Range("D" & Val).Value = Range("D" & Val).Value + 1 ’ incr bin count
End If
Range("A5").Value = Range("A5").Value + 1 ’ increment counter i
Loop
End Sub

© 2015 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 62 CHAPTER 6. MODELING RANDOM DATA & NOISE

Project 6.4 (Random Gaussian noise) Extend Example 13.33 to generate nN = 25 random numbers
Ni for 0 ≤ i ≤ 24 and compute the sample averages. Combine columns B and C to form Ni with a
single formula using the PRNG. Observe how the sample averages change as another set of random
numbers is generated.

(ans:

A B C D
1 N= 4
2
3 i Ni
4 0 4.401
5 1 0.931 Sample Ave=
6 2 5.052 0.509
7 3 4.673
8 4 3.157 Sample SD=
9 5 4.978 3.959
10 6 1.091
11 7 4.841 Sample Var=
12 8 4.215 15.673
13 9 3.565
14 10 2.109
15 11 1.378
16 12 2.788
17 13 5.537
18 14 4.461
19 15 0.966
20 16 4.473
21 17 1.705
22 18 5.163
23 19 1.696
24 20 1.914
25 21 0.483
26 22 8.931
27 23 3.951
28 24 2.801
A B C D
1 N= 4
2
3 i Ni
4 0 =$B$1*NORM.S.INV(RAND())
5 =A4+1 =$B$1*NORM.S.INV(RAND()) Sample Ave=
6 =A5+1 =$B$1*NORM.S.INV(RAND()) =AVERAGE(B4:B28)
7 =A6+1 =$B$1*NORM.S.INV(RAND())
8 =A7+1 =$B$1*NORM.S.INV(RAND()) Sample SD=
9 =A8+1 =$B$1*NORM.S.INV(RAND()) =STDEV.S(B4:B28)
10 =A9+1 =$B$1*NORM.S.INV(RAND())
11 =A10+1 =$B$1*NORM.S.INV(RAND()) Sample Var=
12 =A11+1 =$B$1*NORM.S.INV(RAND()) =VAR(B4:B28)

© 2015 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 6.2. EXCEL PROJECTS 63

Project 6.5 (Noisy digital die-toss signals) Using Example 13.34 as a guide, compose a worksheet
that generates 25 samples of random data corresponding to a die toss that is corrupted by additive
random Gaussian noise with σN = 0.2

(ans: The issues here involve formatting the amplitude in the plot. Because almost all Gaussian values
fall within [-3σ, 3σ] of the mean, for σ = 0.2 the y scale range should be [(1-0.6)=0.4, (6+0.6)=6.6].
A B C D E F G H I J K L M
1 N= 0.2
2
6.4
3 i Di Ni Si = Di+Ni
4 0 2 0.109 2.109 5.9
5 1 2 0.018 1.982
6 2 5 0.126 5.126 5.4
7 3 3 0.268 3.268
8 4 1 0.021 1.021 4.9
9 5 5 0.070 4.930
10 6 4 0.061 4.061 4.4
11 7 5 0.010 5.010
12 8 5 0.179 4.821 3.9
13 9 4 0.078 3.922 Di
14 10 2 0.047 2.047 3.4
15 Si = Di+Ni
11 2 0.158 1.842
16 12 6 0.197 6.197 2.9
17 13 3 0.287 3.287
18 14 5 0.223 4.777 2.4
19 15 3 0.472 2.528
20 16 3 0.034 3.034 1.9
21 17 3 0.220 2.780
22 1.4
18 2 0.180 1.820
23 19 1 0.122 1.122
24 0.9
20 6 0.031 5.969
25 21 1 0.026 0.974
26 0.4
22 5 0.099 4.901
0 5 10 15 20
27 23 5 0.014 4.986
28 24 2 0.022 2.022

A B C D
3 i Di Ni Si = Di+Ni
4 0 =CEILING(6*RAND(),1) =$B$1*NORM.S.INV(RAND()) =B4+C4
5 =A4+1 =CEILING(6*RAND(),1) =$B$1*NORM.S.INV(RAND()) =B5+C5
6 =A5+1 =CEILING(6*RAND(),1) =$B$1*NORM.S.INV(RAND()) =B6+C6
7 =A6+1 =CEILING(6*RAND(),1) =$B$1*NORM.S.INV(RAND()) =B7+C7
8 =A7+1 =CEILING(6*RAND(),1) =$B$1*NORM.S.INV(RAND()) =B8+C8
9 =A8+1 =CEILING(6*RAND(),1) =$B$1*NORM.S.INV(RAND()) =B9+C9
10 =A9+1 =CEILING(6*RAND(),1) =$B$1*NORM.S.INV(RAND()) =B10+C10
11 =A10+1 =CEILING(6*RAND(),1) =$B$1*NORM.S.INV(RAND()) =B11+C11
12 =A11+1 =CEILING(6*RAND(),1) =$B$1*NORM.S.INV(RAND()) =B12+C12

© 2015 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 64 CHAPTER 6. MODELING RANDOM DATA & NOISE

Project 6.6 (Thresholding noisy binary signals) Extend Example 13.35 to form 20 samples of ran-
dom binary data that is corrupted by additive random Gaussian noise with specified σN . Attempt to
determine the data from Si by applying a threshold τ = 0.5 to estimate the data value. Repeatedly
generate new data. Determine the σN value that typically produces 1 error is the 20 values of Si .

(ans:

A B C D E F
1 N= 0.3
2
3 i Bi Ni Si = Bi+Ni ~Bi Error
4 0 1 0.316 0.684 1 0
5 1 1 0.489 0.511 1 0
6 2 1 0.429 0.571 1 0
7 3 1 0.028 1.028 1 0
8 4 0 0.212 0.212 0 0
9 5 0 0.122 0.122 0 0
10 6 0 0.416 0.416 0 0
11 7 1 0.184 0.816 1 0
12 8 0 0.394 0.394 0 0
13 9 0 0.160 0.160 0 0
14 10 1 0.239 0.761 1 0
15 11 1 0.022 0.978 1 0
16 12 1 0.890 0.110 0 1
17 13 1 0.180 0.820 1 0
18 14 1 0.548 1.548 1 0
19 15 1 0.484 1.484 1 0
20 16 1 0.090 0.910 1 0
21 17 1 0.189 0.811 1 0
22 18 0 0.088 0.088 0 0
23 19 1 0.148 1.148 1 0
A B C D E F
3 i Bi Ni Si = Bi+Ni ~Bi Error
4 0 =IF(RAND()<0.5,0,1) = $B$1*NORM.S.INV(RAND()) =B4+C4 =IF(D4<0.5,0,1) =IF(B4<>E4,1,0)
5 =A4+1 =IF(RAND()<0.5,0,1) = $B$1*NORM.S.INV(RAND()) =B5+C5 =IF(D5<0.5,0,1) =IF(B5<>E5,1,0)
6 =A5+1 =IF(RAND()<0.5,0,1) = $B$1*NORM.S.INV(RAND()) =B6+C6 =IF(D6<0.5,0,1) =IF(B6<>E6,1,0)
7 =A6+1 =IF(RAND()<0.5,0,1) = $B$1*NORM.S.INV(RAND()) =B7+C7 =IF(D7<0.5,0,1) =IF(B7<>E7,1,0)
8 =A7+1 =IF(RAND()<0.5,0,1) = $B$1*NORM.S.INV(RAND()) =B8+C8 =IF(D8<0.5,0,1) =IF(B8<>E8,1,0)
9 =A8+1 =IF(RAND()<0.5,0,1) = $B$1*NORM.S.INV(RAND()) =B9+C9 =IF(D9<0.5,0,1) =IF(B9<>E9,1,0)
10 =A9+1 =IF(RAND()<0.5,0,1) = $B$1*NORM.S.INV(RAND()) =B10+C10 =IF(D10<0.5,0,1) =IF(B10<>E10,1,0)

© 2015 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 6.2. EXCEL PROJECTS 65

Project 6.7 (Histogram of noisy binary signals) Using Example 13.32 as a guide generate a histogram
of ten thousand noisy binary signals.

(ans: The random binary data takes on values of +1 and -1 computed in B1. The corresponding signal
is computed in D2. The noise standard deviation is specified in B3. The Sd is convenient for setting the
histogram limits. The Gaussian noise sample is computed in D3, and the signal plus noise is formed in
D4.

The histogram extends from hist min to hist max (= − hist min) and its parameters are computed
in the worksheet. The hist min value in A9 is computed from A and σN as hist min = −A − 4σN )) to
effectively guarantee that it will be less than the minimum Xi value encountered. The number of bins is
specified by the user in B9. The bin width is computed in C9 as −2( hist min)/(# bins-1). The bin number
of an observed value is the row number is the histogram display and is computed as rounded((Xi value
minus hist min)/bin width) +2 in C6 as
=ROUND((D4-A9)/C9,0)+2
This allows the verification of the bin number for each manual observation (produce by F9).
A B C D E F G
1 D= 1 bin count
2 A= 10 Si = 10.00 18 0
3 N= 2 Ni= 0.42 16 8
4 Xi= Si+Ni 9.58 14 40
5 nT= i= bin#= 12 112
6 1000 1000 6 10 200
7 8 113
8 hist min = # bins= width= 6 26
9 18 19 2 4 3
10 2 0
11 0 0
12 2 0
13 Hist 4 3
14 6 32
15 8 114
16 10 215
17 12 101
18 14 33
19 16 0
20 18 0
A B C D
1 D= =2*ROUND(RAND(),0) 1
2 A= 10 Si = =B1*B2
3 N= 2 Ni = =B3*NORM.S.INV(RAND())
4 Xi= Si+Ni =D2+D3
5 nT= i= bin#=
6 1000 1000 =ROUND((D4 A9)/C9,0)+2
7
8 hist min = # bins= width=
9 = ROUND(B2+4*B3,0) 19 = ROUND(2*A9/(B9 1),0)

Performing most of the computations in the worksheet allows the following VBA code to be relatively
simple.

© 2015 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 66 CHAPTER 6. MODELING RANDOM DATA & NOISE

Sub hist()
Dim Val As Integer
Range("B6").Value = 0 ’ Reset counter i
For Val = 1 To Range("B9").Value ’ Set Hist Bin Vals
Range("F" & 1 + Val).Value = Range("A9").Value + (Val - 1) * Range("C9").Value
Range("G" & 1 + Val).Value = 0 ’ Initialize counts
Next
Do While Range("B6").Value < Range("A6").Value ’ loop for nT times
Val = Round((Range("D4").Value - Range("A9").Value) / Range("C9").Value) + 2 ’bin value
If Val > 1 Then ’ if valid bin value
Range("G" & Val).Value = Range("G" & Val).Value + 1 ’ incr bin count
End If
Range("B6").Value = Range("B6").Value + 1 ’ increment counter i
Loop
End Sub

© 2015 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.