CourseSmart - Instructors - Print

http://instructors . courses mart. coml pri nt?xml i d =9780470252543/100 ...

lJ~~LD<tm.~: ~.i?9.~.R.~I .. [email protected]~~ ~QQ~;Pd~~.iRI.~~ .. .Q.f..(;:Qm.m.~!1.ig.\ig!1~, .. Q.\h .. ~Qi.\.iQfI P~.g.~.:.J . .QQ

No part of any book Terms of SeN ice is

100 Cnapter 2 • Signal and linear System Analysis

Problems

any means without the publisher's prior permission. Use (other than qualified fair use) in violation of the law or the full exte nt of the law.

Section 2.1

2.1. Sketch the single-sided and double-sided amplitude and phase spectra of the following signals:

a. xa(t) = lOcos(4m + 17'j8)

+ 6 sin(817't + 317' /4). b. Xh( t) = S cos( 2m + 17' /3)

+ 4cos(617'f + 17'/4).

C. xc(t) = 2 sin(4m+7TjS) +12sin(I07T£).

2.2. A signal has the double-sided amplitude and phase spectra shown in Figure 2.34. Write a time-domain expre ssion tor the signal.

2.3. The sum of two or more sinusoids may Of may not be periodic depending on the relationship of their separate frequencies. For the sum of two sinusoids, let the frequencies of the individual terms bell and/i. respectively. For the sum to be periodic.jj andh must be commensurable: i.e .. there must be a number/o contained in each an integral number of times. Thus. if.(o is the largest such number.

/: = ntfo and [: = lr2fo

where nl and n2 are integers: Io is the fundamental frequency. Which of the signals given below are periodic') Find the periods of those that are periodic.

a. XI(l) = 2cos{2t)+4sin(6m).

b. X2(t) = COS(617I) + 7cos(307Tt).

c. :q(t) = COS(47T1) +9sin(2l7TI).

d. X4(t) = 2cos(4171) + 5cos(67Tt) +6sin(177Tt). 2.4. Sketch the single-sided and double-sided umplitude and phase spectra of

i·'..mplitude

a. xa(t) = 5cos(1217't-7Ti6).

b. Xb(l) = 3 sin(127T1) + 4 cos{l617l).

c. x,.(t) = 4 C05(S17't) cos( 1217't).

(Him: usc an appropriate trigonometric identity to write as the sum of cosincs.)

d. xAI) = Ssin(217f)cos2(5171).

i Hint: usc appropriate trigonometric identities.) 2.5.

a. Show that the functioni5,U) sketched in Figure 2.4

(b) has unity area,

b. Show that

8,it) = €~ f e-'/'uU)

has unity area. Sketch this function for If = L t, and ;t. Comment on its suitability as an approximation forthe unit impulse function.

c. Show that a suitable approximation for the unit impulse function as e ~ 0 is given by

\ e-I (l-1.1). itl:S e

is.(!) = e

O. otherwise

2.6. Usc the properties of the unit impulse function given after (2.14) to evaluate the following relations.

a .. C!t2 + sin(27T1)]6(2t-5)dt.

b. f~~~ (/2 + 1)[L;~~~B(I-5n)ldl.

tNore: ! O ~ mean s jus t to the right of 10: - to - mean s just to the left of - 10.)

Figure 2.34

1 of 10

10/4/20101:14 PM

CourseSmart - Instructors - Print

http://instructors.coursesmart. com/print?xml i d =97 804 7025 2543/100 ...

.v~~L~."tl).~: ~.i;g.".!?~I .. ~.Q@J~~ ~Q"!~:. . .m~"!2!.~.~ .. .Q.f.(:Q'.T'.'.T'..~Qi.<;~TiQ~~, §.\h.~<:I.iti.Q!'. P".g.~: . ..1..QJ.

No part of any book may be rep rod uced or transmitted by any means without the pub lisher's prior permission. Use (other than qualified fair use) in vio lation of the law or Terms of Service is prohibited. Vio lators w ill be pro sec uted to the full extent of the law.

, __ ,,_, __ , __ ,_ ,_ .,_,_co- .,_,_m_" " " " " •• " c ••• com ." •• om m "m •• " •••• "." ...

e. !OS(t) + A df,(t)/dt = Bf,(r) + 5 df,(r)/dl: findAandB.

d • .hI! [e-41r' + tan(I07TI)JS(3t+ 6)dt.

e. J:'M [eos{87Tt) +e-2'J W8{r - 2)ldP] dt .

2.7. Which of the following signals arc periodic and which are aperiodic? Find the periods of those which arc periodic. Sketch all signals.

a. Xl(t) = cos(S7Tl) + sin(77T/).

b. X2(/) = L;=oA(t-2n). c, X3(t) = L:_~A(I~2n).

d . .1."4(/) = sin(3t) + COS(27TI).

e. xs(t) = I::-~ n(I--3n).

f. X6(r) = L;=on(r-·3n).

2.8. Write the signa! x{t) = sin(67T1) + 2cos(107TI) as

a. The real part of a sum of rotating phasors.

b. A sum of rotating phasors plus their complex conjugates.

c. From your results in parts (a) and (b). sketch the single-sided and double-sided amplitude and phase spectra ofx(t).

Section 2.2

2.9. Find the normalized power for each signal below that is a power signal and the normalized energy for each signal that is an energy signal. If a signal is neither a power signal nor an energy signal. so designate it. Sketch each signal (a is a positive constant),

a . .'<"1(1) = 2cos(47T1+27Ti3).

b. X2(t) = C""U(l).

c. XJ (t) = e""'u( -t).

d. X4(t) = (0:'2 + p.r~I/2 <

e. Xs (t) = r'd'i.

f, X,; = ,.,-'''u(r)-e-''(I-I}U(I-l).

2.10. Classify each of the following signals as an energy signal or a power signal by calculat ing the energy E or the power P V1. g. W'. and r arc positive constants).

2. Ai .in(;,)I.).. f.I) I.

b. M/Jr+jl.j = R.

c. Attr'!'u(l).

d. n(t/!)+n(I/2~).

2.11. Sketch each of the following periodic waveforms and compute their average powers.

a. XIU) = L:~M ni(t-6n)/3].

2 of 10

b.:Q(I) = L;~_~A[(t~5n)/21.

c. X3\t) = I::=_MAi(t-3n)/2)u{t-3n).

d. X4(t) = 2 i>in(57Tt) COS(S7TI).

(Hint: use an appropriate trigonometric idcnty to simplify. 2.12. For each of the following signals. determine both the normalized energy and power. (Note: () and co arc possible answers.)

a. XI (t) = 6,,(-3+/4")'u(l).

b. X2(1) = n!(r-3 1/2J + nl{l-3)/6J.

c. X3(t) = 7ejfm'u(t).

d. -'4(1) = 2cos(47Tl).

2.13. Show that the following arc energy signals. Sketch each signal

a. XI(/) = n(I/!2)cos{67T1)

b. X2(1) = Clti/3

C. x3(t) = 211(1)-211(t-8)

d. X4(1) = roo u(i.)di.-2 f:10 uU)di. + r::,20 1I(1.)dl-

(Him: Consider the integral of a step function.)

Section 2.3 2.14.

a. Fill in the steps for obtaining (2.33) from (2.32). b_ Obtain (2.34) from (2.33).

c. Given the set of orthogonal functions

.I.n(l) = n (4!1-(2J1T-l)T /8)).

'" n = [,2.3.4

sketch and dimension accurately these functions. d. Approximate the ramp signal

I In(I~T.!2) x/)="T -T-

by II generalized Fourier series usi ng thi s set. e. Do the sa me for the set

f. Compute the integral-squared error for both part (b) and part (c), What do you conclude about the dependence of EN on N?

10/412010 1: 14 PM

CourseSmart - Instructors - Print

http://instructors.coursesmart.com/pri nt?xml id=97804 702525431 100 ...

VgL.narng:L,I?o:I"b~I.M.9@I~.s ~9.Q.~: f'.rj!1:~i.2.I.~.~.Qf..<;:9."'.rn.y.Qi~.a-'-IQ!1:s., . .§.;.h .. ~.t:I!\.IQD. P"gg:.J.Q.f.

No part of any book any means without the publisher's prior permission. Use (other than qualified fair use) in violation of the law or

Terms of Service Is the full extent of the law.

102 Chapter 2 • Si9na[ and U.,.;,ar S)I5tem Analysis

Section 2.4

2.15. Using the uniqueness property of the Fourier series. find exponential Fourier series for the following signals (10 is an arbitrary frequency):

a. Xl(/) = 5in2(21ifol).

b. -"l{t) = oos(21ifo/) + sin( 4r.:fot). c, X3 (I) = si n( 4r.fo t) cost 41ifot) .

d. X4( t) = cos" (2r.fof).

Hint: Use appropriate trigonometric identities and Euler's theorem.

2.16. Expand the signal x(t) = 2? in a complex exponential Fourier series over the interval I rl S 2. Sketch the signal to which the Fourier series converges for all I.

2.17. If Xn = IX. lexp f/L&.] are the Fourier coefficients of a real signal. x( t). fifl in all the steps to show that:

a. IX.I = iX_ni and!Xn = -!'X-IJ....

b. Xn is a real. even function of n for x(t) even.

c. X" is imaginary and an odd function of n for x(t) odd.

d. X(I) = -x(/ + To/2) (half wave odd symmetry! implies that Xn = 0, n even.

2.18. Obtain thecomplexexpcnential Fourier series coefficients fOfthe(a)pul se train. (b) hal f-rectificd si newave. (c) full-rectified sine wave. and (d) triangular waveform as given in Table 2.1.

2.19. Find the ratio of the power contained in a pulse train for il!(o I S T~ I to the total power for each of the following cases:

a. r;ITo = t.

b. r/To=t

c. TiTo = 10.

d. r,!To =10·

Hint; You can save work by noting the spectra arc even aboutj' = O.

x)lI

!

-r-n.u.

0: To 2To

(u)

2.20.

a. If x( t) has the Fourier series

x(t) = L x""fl.1I'nfo!

I;:::::-'X'I

and Y(l} = X(l-lo). show that

where the Yn are the Fourier coefficients for y(t).

h. Verify the theorem proved in part (a) by examining the Fourier coefficients for x(t) = cos(WOt) and y(t) = sin(wol).

liint; What delay. to. will convert a cosine intoa sine. Usethc uniqueness property [0 write down the corresponding Pourier series.

2.21. Use the Fourier series expansions of periodic square wave and triangular wave signals to find the sum of the following series:

a.l-t+*-++··

b. 1 + ~ + ~ + $- + ....

Hint: Write down the Fourier series in each ease and evaluate it for a particular. appropriately chosen value of t.

2.22. Using the results given in Table 2 [ for the Fourier coefficients of a pulse train. plot the double-sided amplitude and phase spectra for the waveforms shown in Figure 2.35. Hint: Note that Xh(t) = -Xa{t) + A. How is a sign change and DC level shift manifested in the spectrum of the waveform'!

2.23.

a. Plot the single-sided and double-sided amplitude and phase spectra of the square wave shown in Figure 2.36(a).

b. Obtain an expression relating the complex exponential Fourier series coefficients of the triangular

·"1,(1)

'Rhnn

0'

Figure 2.35

3 of 10

10/4/20101:14 PM

CourseSmart - Instructors - Print

http://instructors.coursesmart.coll1}print?xml id=97804 70252543/100 ...

Us. r name: Lizdabel Mo rales m ..,. • .., .., ~Q9.K:PJin~iple~ .QtC;Qmmvni.<;.ali9.n~,j;l:lb.~9j;iQn ea9.~.;.J .. Q}

No part of any book may be reproduced or transmitted by any means without the publisher's prior permission. Use (other than qualified fair use) in violation of the law or Terms of Service is prohibited, Violators w ill be prosec uted to the full extent of the law.

Problems 103

.\J.tl I

D~b D

-To y-A U ~

(a)

Figure 2.36

waveform shown in Figure 2.36(b) and those of xa(!) shown in Figure 2.35(a).

Hint: Note that xu(t) = Kld'(I,(t)/dtj. where K is an appropriate scale change.

c. Plot the double-sided amplitude and phase spectra for Xb{t).

Section 2.5

2.24, Sketch each signal given below and find its Fourier transform. Plot the amplitude and phase spectra of each signal (A and r are positive constants).

a. -"1(1) = Acxp( -1!r)tI(I). h. X2(f) = Aexp(t/r)II(-t).

c. -"3(f) = .\'!(r)-x2(/).

d. X4(/) = Xl(t)+X2(t). Docs it cheek with the answer found using Fourier transform tables')

2.25.

a, Usc the Fourier transform of

xU) = exp( -o:l)II([)~exp(at)lI( -I)

where a > 0 to lind the Fourier transform of the signum function defined as

{l.

sgnl = _!.

(Hint: Take the limit as a ~O of the Fourier transform found.)

b. Usc the result above and [he relation tI( t) = ! [sgn f + 1] to find the Fourier transform of the unit step.

c. Usc the integration theorem and the Fourier transform of the unit impulse function to find the Fourier transform of the unit step. Compare the result with part (b).

2,26. Using only the Fourier transform of the unit impulse function and the differentiation theorem, find the Fourier transforms of the signals shown in Figure 2.37.

4 of 10

(b)

2.1.7.

a, Write [he signals of Figure 2.37 as the linear combination of two delayed triangular functions. That is, write xa(r) = GtA{(l-ll )ITl) +ozA((t-fz)/Tz) by finding appropriate values for at, t12, [I • f2, Tt, and Tz. Do similar expressions for all four signals shown in Figure 2.37.

b. Given the Fourier transform pair A(l)"""" sinc2f, find their Fourier transforms using the superposition. scale change. and time delay theorems. Compare your results with the answers obtained in Problem 2.26.

2.28.

a, Given n (I) ......,. sine j, find the Fourier transforms of the following signals using the frequency translation followed by the time delay theorem.

L Xl(t) = n(t-l)cxp!j41T(I-l)), ii. -"z([) =n(t+ l)exp!J4?T(t+ I)].

b. Repeal [he above, but now applying the lime delay followed by the frequency translation theorem.

2.29. By applying appropriate theorems and using the signals defined in Problem 2.28, find Fourier transform, of the following signals:

a. -"a{l) = !Xt{I) -+ ~Xl( ~r).

b. x,,([) = tXz(t) + tX2( -f).

2_~~. lise the scale change and time delay theorems along with the transform pair, n(I)"""" sine f. sine I"""" ncr). A(t),___,. sine2r. and sinc2t - A(f) to find Fourier transforms of the following:

a. xc(t) = nl(I-·l )/2].

h. X/,(I) = 2sincI2(r- 1)1.

c. x<(I) = AI(t~2)/8].

d. X,J(I) = 5in2[(1-3)/4].

10/4/20101:14 PM

CourseSmart - Instructors - Print

http://instructors.coursesmart.comiprint?xml i d=97804 70252543/100 ...

~()Qk;~ri!,Cip!~.~ .. g.f..r;,()_!,,[1J.~!"!.i.<:.i'.\i().n.'~.' .. .§.'.h.gt!i,jQ!' E~.g.~.:.J .. Q.~

any means without the publisher's prior permission. Use (other than qualified fair use) in violation of the law or the full extent of the law.

104 Chapter 2 • Signal and Linear Sy;tem Analy,i,

t, s

2r-----------~----~1-------,:

I :

I ! i

,I I ! I I i

1.5 ~-_t----1-----;-----:-----~--~

r . ~ ~ i : I

I I 1 , I ~ i

~ 1 ~-1---- '----~---- I-----:--~

i

I !

~-r--i

I I

OL-J ~ L ~ '~

o

2 !,;;

Figure 2.37

r:f 0

I I

05 -~-- --:-----:-----:--

~ I I t

~ I I t

I I I

! I I I

-~~-~~~-- ~---~----

I I I

I I I

I I I

'I I

f I I I

-j----,----j-- -1--

f ! I I

I I I

1 ~ : :

_1~1~ ~ __ ~ __ ~ l_~

-05

I}

4

2 t,«

2.-----------------------,--,

" ,

I I ~ r I I

1.5 r' -4----J-----:-----:-----:--

I , r I I

I ~ ! I I

~ 1 t-~----: ----~---- :----~--

I r I I

I r I I

I r! I

0.5 ~~__ r; _~ __

I , I

231. Without actually computing them. but using appropriam sketches. tell if the Fourier transforms of the signals given below are real. imaginary. or neither; even, odd. or neither. Give your reasoning in each case,

a. Xl(£) =D(r+1/2)~D(l--1/2).

b. X2(l) = nU/2) +n{l).

c. X3(t) = sin(2r.t)D(t).

d. X4(t) = sin(2m+r./4)n(t). c. -'s(t) = cos(27T1)D(I).

f. -'6(t) = I/[! + (t/5)4].

2.32. Using the sifting property of the della function. find the Fourier transforms of the signals given below. Discuss how any symmctry properties a given signal may have affect its Fourier transform in terms of being real or purely imaginary.

a. X,(l) = 5{1 + 4) + 35(1) + 8(r~4). h. :dl) = 28(1 ·18)--<Z8(/·-8).

c • .\"3(t) = L!~ (n2 + [)0(/-2n). (Hint: Write out the terms for this signal.)

5 of 10

r. S

d .. \"4(1) = I:;~ .. 2!,2S(r-2n)

(flint: Write out the terms for this signal.)

2.33. Find and plot the energy spectral densities of the following signals. Dimension your plots fully. Usc appropriate Fourier transform pain and theorems,

a .. \·l(t) =2c31'i.

b • .\"2(1) = 20 sinc(30t). c .. \·3(t) = 4n(5/).

d. X4(1) = 4D(51)cos(401Tl).

234. Evaluate the following integral" using Rayleigh's energy theorem {Parseval's theorem for Fourier transforms).

a. [1 = J,"~ !""+~;"irj'

(Hint: Consider the Fourier transform of cxp( -ta)u(t)).

b. I: =.C sinc1(Tf)d(.

c. h = J~~ r ".} ~ n21--'

[iT"" .. '"~!< J

d. 14 = f'M sinc4(T.f)d(.

235. Obtain. and sketch the convolutions of the followi ng signals.

10/4/20101:14 PM

CourseSmart - Instructors - Print

http://instructors.coursesmart.com/print?xmlici=97804 7025 2543/100 ...

U$~rnittl1".:LiZcl~.fJ~I.M<:>.@J~L ?<?9.~: .. ?ti.Q~i.pl.~~.QJJ,:<?m.tl1.~.l)i., .. "J.i9.!)~, .. §th.l'.g..iti<:>D. P.?.9.~:.J . .Q.?.

No part of any book any means without the publisher's prior permission. Use (other than qualified fair use) in violation of the law or

Terms of Setvice is the full extent of the law.

a. Ydt) = e~afU(t) " n(t~T),

cr and'! positive constan ts,

b. Y2(t) = !n(t/2) + fl(t)] " nit).

c. )'3 (I) = r"I'1 * n(t). ex > O.

d. Y4(1) =x(t) " u(t). where X(I) is any energy signal [you will have to assume a particular form for X(I) to sketch this one, but obtain the general result before doing so],

236. Obtain thc Fourier transforms of thc signals YI(t), Y2(t). and YJ(t) in Problcm 2.35 using the convolution theorem of Fourier transforms.

2.37. Given the following signals. suppose that all energy spectral components outside the bandwidth If! 5 W are removed by an ideal filter. while all energy spectral components within this bandwidth are kept. Find the ratio of output cnergy to total energy in each case. (cr. /3. and T are positive constants.)

a. Xl(t) = e-ttfu(t).

b •. \'2(t) = n(t/T) (requires numerical integration). c. X3(t) = e-"Iu(t)~e-{Jru(t) (f3 = 2a).

a. Find the Fourier transform of the cosine pulse

. (2t)

x(t) = An To cos(rlQ/)

where Wo = f.'. Express your answer in terms of a sum of sine functions. Provide MATLAB plots 01'.\'( t) and XU) [note that XU) is real].

a. Obtain the Fourier transform of the raised cosine pulse

y(t) =~AnG~)!1 + cos(2WoI)]

Provide MATLAB plots of yet) and Yff) [note that YCf) is real], Compare with part (a).

c. Use (2.151) with the result of part (a) to find the Fourier transform of the half-rectified cosine wave.

239. Provide plots of the foHowing functions of time and find their Fourier transforms. Tell which Fourier transforms should be real and even functions of.r and which ones should be imaginary and odd functions off. Do your results bear this om?

a. XI (I) = A(~-) + n (t).

b. \'2(/) = n(t/2)~A(I).

c. X3(t) = n(t+ tl -n(t·· t)·

d. X4(1) = A(I-·l)-A(I+ 1).

6 of 10

Problems 105

Section 2.6 2.40.

a. Obtain the time-average autocorrelation function ofx(t) = 3 + 6cos(20171) +3 sin(2017t).

(Hint: Combine the cosine and sine terms into a single cosine with a phase anglc.)

b. Obtain the power spectral density of the signal of part (a). What is its total average power?

2.41. Find the power spectral densities and average powers of the following signals,

a. X!(t) = 2 cos (20 7TI +17/3).

b. X2(1) = 3 sin (30171).

c . .\'3(1) = 5 sin(I0l7l-17/6).

d. X4(1) = 3 Sin(30171) + 5 sin( 10'171-'17 /6).

2.42. Find the autocorrelation fu notions of the signal s having thc following power spectral densities. Also give their a verage powers,

a. 5l en = 4O(f ~ 15) +40(.f + 15).

b. 52(fJ = 90(/-20) + 90(.f + 20).

c. 53(f) = 160(.f~5)~~ I 6:S(f + 5).

d. 54(/) = 90(.f~20) + 90(f + 20)

+ 160(f -5) + 16:S(f + 5).

2.43. By applying the properties of the autocorrelation function. determine whether the following arc acceptable for autocorrelation functions. In each case. tell why or why not.

a. RI(r) = 2cos(Wm) + cos(30'17o).

b. R2(r) = 1 + 3cos(30m:).

c. R3(T) = 3eos(20m 1- 17/3).

d. ~(r) = 4A(r/2).

e. Rs(t) = 3n{t/6).

f. R6(T) = 2sin(IOm).

2.44. Find (he autocorrelation functions corresponding (0 the following signals:

a. x! (I) = 2eos( IOrrl + 17/3).

b. X2{l) = 2sin(IOm +17/3).

c . .\'3(1) = Re(3 exp(jl (171) + 4j exp(j 10'171). d. X4(t) =Xl(t)+X2(t),

2.45. Show that the R( r) of Example 2.20 has the Fourier transform given there. Plot the power spectral density,

Section 2.7

2.4(,. A system is governed by the differential equation (a. b. and c arc nonnegative constants)

10/412010 1:14PM

CourseSmart - Instructors - Print

http://instructors.coursesmart.comJprint?xmlid=97804702525431100 ...

.W.~.~LD~.m.~:.~!~9~.logl.M<?gJ~~ ~ggL.!'linc:iples()Lc:()rnm~ni<:~.ti{)D.S.,.~th.~Q.it.i9.n ~~gg: .. !QEi

~:~~~?Je';:;rc~?s°k

any means without the publisher's prior permission. Use (other than qualified fair use) in violation of the law or t he full ext. nt of the law.

106 Chapt",2. Sign"l and Linear SY"tf'm Analy~rs

dv dx

...:.. +av =h- +cx

dt e dt

a. Find H( f).

b. Find and plot lfi (.f) I and I H cn for c = O.

c. Find and plot IHer)! and IJflfl for b = O.

2.47. For each of the following transfer functions, determine the unit impulse response of the system.

a. HI (f) = (5+)2<'Jfi

b. H2(f) = (5~j{<'Jf)

(Hint: Use long division first.) c. H3(fl = (5 '~~;;1l1T

d H (f) I_c~~<f

• 4. = (IT'7iWi)'

2.48. A filter has frequency-response function N(f) = nU/2B) and inpu!X(t) = 2W sinc(2Wt).

a. Find the output y(f) for W < B.

b. Find the output )'(1) for W> B.

c. In which case docs the output suffer distortion'!

What influenced your answer?

2.49. A second-order active bandpass filter (BPF). known as a bandpass Sallen-Key circuit is shown in Figure 2.38.

a. Show that the frequency-response function of this filter is given by

I' ~ (KW(}j!2)(jw)

li,./fJJ) - , I 'Q)I') " t» = 2'1't[

-0>" + 'Wfi! ./w + W(j

R

'---1 -~W,------'I

1', R t C

~(-( --,------!

!

~c R I

~

Input

Figure 2,38

70f10

where

Wo = '/2{RC)'!

j2 Q= 4-K

K=!+Ra R;,

b. Plot IH(I)I,

c. Show that the 3-dB bandwidth in hertz of the filter can be expressed as B = fo/Q. wherefn = Wo/27l.

d. Design a BPF using this circuit with center frequency [0 = 1000 Hz and 3-dB bandwidth of 30() Hz. Find values of Ra,R",R. and C that will give these desired specifications,

2.50. For the two circuits shown in Figure 2.39. determine NU) and h(t). Sketch accurately the amplitude and phase responses. Plot the amplitude response in decibels. Usc a logarithmic frequency axis,

2.51. Using the Paley-Wiener criterion. show that

is not a suitable amplitude response for a causal. linear time-invariant filter.

2.52. Determine whether the filters with impulse responses given below are BIBO stable.

3. IIdf) = cxp(,al) COS(21ifof)U(t).

b. h2{f) = cos(27Tf~{)[I(f).

c. !t3(t) = c1u(f-· I),

Output

10/4/20101:14 PM

CourseSmart - Instructors - Print

http://instructors .coursesmart.comlprint?xml id=97804 70252543/100 ...

No part of any book may be reproduced or transmitted by any means without the publisher's prior permission. Use (other than qualified fair use) in violation of the law Or Terms of Service is prohibited. Violators Will be prosecuted to the full extent of the law.

Problems 107

+

+

+

RZ

y(l)

xO)

xU)

L

L

)it)

Figure 2.39

253. Given a filter with frequency-response function ~

flU) = 4+.i(2'1f{)

and input x(t) = e-3'u(I). obtain and plot accurately the energy spectral densities of the input and output.

2.54. A fi Iter with frequency-response function

HU) = 3D ({6)

has. as an input. a half-rectified cosine waveform of fundamental frequency lO Hz. Determine the output of the filter.

255. Another definition of bandwidth for a signal is the 90% energy containment bandwidth. For a signal with energy spectral density GU) = !XU)11• it is given by Bw in the relation

rn., r!ho

J GU) {(( = 2 J GU) {((

-B", 0

[~ GU) dl" = 2 [ GU)dl"

ObtainB'IO forthe following signals ifit is defined.If it is not dctined for a particular signal. stale why it is nOL

a. Xr(/) = e·-"'u(t). where 0: is a positive constant.

b. X2(t) = 2Wsinc(2Wt).

e. X3(t) = D(ljT) (requires numerical integration). 2.56. An ideal quadrature phase shifter has

{(dtr!2. I> () HU) = e+J"j2. f < ()

Find the outputs for the following inputs:

a. Xl (I) = exp(flOOl7I).

b. X2{t) = cost !OOl7/). e . .'1'3(1) = sin] 10017/). d. X4(t) = D(lj2).

257. A filter has amplitude response and phase shift shown in Figure 2.40. Find the output for each of the inputs given below. For which cases is the transmission distortionlcss? Tell what type of distortion is imposed for the. others.

a. Xdl) = cos(4817t) +5cos(126r.-t).

b •. \'2(1) = cos(126?l't) + O.S cos(17017/). C • .'1'3(1) = cost 126m) + 3cos(144m). d. ):4(1) = cos(101T/) +4eos(50m).

2.5S. Determine and accurately plot, on the same set of axes. the group delay and the phase delay for the systems with u nil impulse responses:

a. 111(1) = 3e-~III(lj.

b. 112(1) = 5e-3(u(l)-2e~5'u{l).

/!!J1)

4

i

~fiH7.)

~ . .

-r -_,_--

~r- ~

- _ __j_IO-()-_-5:-0--(C-) if---:"50"---t-':()IC-) =! (Hr.)

Figure2AO

8ofl0

10/4/20101:14 PM

CourseSmart - Instructors - Print

http://instructors.coursesmart.com/print?xmIid=97804 70252543/100 ...

,I,!~ecn,\'!'~: P~g.~~~.I.~.Q@J~.s.............................................. . ~9.g.~: .. E~.'!~i.R.I.~ .. S .. (?.f..<:;9.rllmHQis:~.t.i9".s,§thEclit.iQn .. P'~g~:J.Q§

No part of any book may be reproduced or transmitted by any means without the publishers prior permission. Use (other than qualified fair use) in violation of the law or Terms of Service is prohibited. V io lators w ill be prosec uted to the full extent of the law .

............ , , " , , , .. "'"."''''"."."" .. , .. ,." .. ,."."."."."." .. ,."." .. ,.,,.,,".".,,'"'''''' " " "'" "'" "'".,, , .. ,., .. , .. "." ,. "., .. ,. - -- ""',."." "."."." .. "" ".".,." .. , .. , .. ,.".,, .

108 Chapter 2 • Signal and Linear System Analysis

2.59. A system has the frequency-response function 4'IT + '27Tf

HU) = .I.

. 8'IT +J27Tf

Determine and accurately plot the group delay and the phase delay.

2.60. The nonlinear system defined by

y(t) = x(l) +0.1 x2(t)

has an input signal with the bandpass spectrum

X(/) = 4[[ (r~2()) + 4[[ (f' :20)

Sketch the spectrum of the output, labeling all important frequencies and amplitudes.

2.61.

a. Consider a nonlinear device with the transfer characteristic y(/) = x(t) +O.lx3(t). The frequency of the input signal x(t) = cos (2(x)01TI) is to be tripled by passing the signal through the nonlinearity and then through a second-order BPF with a frequency response function approximated by

H(f' ~ 1 + !

. I - I +/2QU ~300:}) I + J2Q(f + 300:))

Neglecting negative frequency contributions, compute. in terms of the parameter Q, the total harmonic distort ion (THDl at the tripler output, defined as

THD = total power in all output distortion term~ x 100% power in desired output component

Note that the desired output component in this case is the third harmonic of the input frequency.

h. Find the minimnm value of Q that will result in THD :5 o.co: %.

2.62. A nonlinear device has y(/) = ao + alx(t) +(/2.,2(1) +a3x1{ti.lfx(t) = COS(Wlt)+ COS(W2}I.list all the frequency components present in y(ti. Discuss the use of this device as a frequency multiplier.

2.63. Find the impulse respom.e of an ideal highpass filter with the frequency response function

2.64. Verify the pulscwidth-bandwidth relationship of (2.250) for the foll owi ng signals. S ketch each signal and its spectrum.

9 of 10

a. X(l) =A exp] ~t2i272) (Gaussian pulse) h. X{I) =Aexp(~alll), a> () (double-sided exponential).

2.65.

a. Show that the frequency response function of a second -order Butterworth filter is

HI{')= fl

,. If +/.Jifl{ ----.(2

where}:" is the 3-dB frequency in hertz.

h. Find an expression for the groupdelay of this filter.

Plot the group delay as a function off I.h.

c. Given that the step response for a second-order B utterworth fi Iter is

v,(t) = {!~exp(~ 2-rr.~!) r cos (271'-0,/)

. . v'2 L v'2

+ SinCizt) ]}U(l)

where 11(1) is the unit step function. find the [()% to 9()% risetimc in terms of.6.

Section 2.8

2.66. A sinusoidal signal of frequency 1 Hz is to be sampled periodically.

a. Find the maximum allowable lime interval between samples.

b. Samples are taken at j-s intervals (i.e .. at a rate of .f~ = 3 sps), Construct a plot of the sampled signal spectrum that illustrates that this is an acceptable sampling rate to allow recovery of the original sinusoid.

e. The samples are spaced ~ s apart Construct a plot of the sampled signal spectrum that shows what the recovered signal will be if the samples arc passed through a lowpass filter such that onl y the lowest freq uc ncy spectra I lines are passed.

2.67. A flat-top sampler can be represented a, the block diagram of Figure 2.41.

:I. Assuming T « 7'." sketch the output tor a typical x(l).

h. Find the spectrum ofthe oU!PU!, ycn, in tenus of the spectrum of the input. X(I). Determine relationship bctween r and T, required to minimize distortion ill the recovered waveform?

2.68. Figure 2.42 illustrates so-called zero-order-hold reconstruction.

10/4/2010 1:14 PM

CourseSmart - Instructors - Print

http://instructors.coursesmart.com/print?xmlid=97804 70252543/100 ...

.lJ.sgc.nil"'e.: li~d~lJgIMg@le~ ~9 . .9.t.: ?tj.Q~.i.QI.~§.9.L~9 .. m"'.~D.iS."'~.i9.DS., .. ii:Ih .. srj.i\.i<:>n p."9.~: .. J.Q.9

No part of any book may be reproduced or transmitted by any means without the publisher's pear permission. Use (other than qualified fair use) in violation of the law or Terms of Service IS prohibited. Violators w ill be prosec uted to the full extent of the law.

. " "." ..

Problems 1 09

Figure 2.41

Figure 2.42

a. Sketch Y(I) for a typical x(r). Under what conditions is y( t) a good approximation to x(/)?

h. Find the spectrum ofy(l) in terms of the spectrum of x(t). Discuss the approximation of y(t) to x(l) in terms of frequency-domain arguments,

2.69. Determine the range of permissible cutoff frequcncies for the ideal lowpass filter used to reconstruct the signal

X(E) = lOcos(600m) eos2(2400m)

which is sampled at 6000 samples pcrsecond. SketchX(I) and Xs(f). Find the minimum allowable s.ampling frequency.

2.70. Given the bandpass signa! spectrum shown in Figure 2.43. sketch spectra for the following sampling rates_(, and indicate which ones arc suitable: (a) 2B. (b) 2.SB. (e) 3B. (d) 4B. (e) SR. (06B.

XI!)

~;~---:;?1

-38 -2B -8 0 i B 2B 38 ,l"(Hz)

Figure 2.43

Section 2.9

2.71. Using appropriate Fourier transform theorems and pairs. express the spectrum Y(f) of

yet) = x(t)cos(wotJ + xC!) sin(u)or)

in terms of the spectrum XU) of X(I). where XU) is lowpass with bandwidth

B<fi. =:::!2.

." 2'tf

Sketch f(f) for a typical XU).

10 of 10

yU)

2.72. Show that X(l) and x(t) are orthogonal for the following signals (wo > 0):

a. Xa(l) = sin(wot)

b. Xb{l) = 2eos(w{)I)"1 sin(WoE) cos(2wot)

c. x<(I) = A expUWoI)

2.73. Assume that the Fourier transform of X(/) 1.<; real and has the shape shown in Figure 2,44, Determine and plot the spectrum of each of the following signal«:

a. XI (E) = rX(E) + ~j.>:(r).

b. -'"l(/) = [ix(/) + iiX(t) ]ef2""!", Io» w. c .. >:J(E) = [tX(/) + ii.\'(t) ]tf2,,"wr.

d. X4(1) = [jx(l)~ ~i x(t)]ei"wr.

IX(j)

I !

L

---~L.w---__jO-------"M-r --f

Figure 2.44

2.74. Consider the signal

X(E) = 2W sinc(2WI) coS(2r.:(ol). /G> W

a. Obtain and sketch the spectrum of x,,( t) = x(1) +JX(I).

b. Obtain and sketch the spectrum of xn(tl '" x«()~jx(t).

c. Obtain and sketch the spectrum of the complex envelope ,>:{f). where the complex envelope is defined by (2.310),

d. Find the complex envelope .>:(/).

10/4/2010 1: 14 PM