Chapter 2 ee 2.1+1 Let us denote the signal in question by 9(t) and its energy by Ey. For parts (a) and (b) 2 2 aintrat ad [ae— 3 [cos rtdt= ton aj, “2 2 if” if” wan} [Cand [Uomanrse [fF esinn? a wafE fad [ae] - [x +0] = 48 Sign change and time shift do not affect the signal energy. Doubling the signal quadruples its energy. In the name way we can show that the energy of g(t) is k*Ey. 242 (a) Kee fitur=2 By = fi (raet f2(-1)Pa (a) Therefore Exsy = Ex~ Ey ooten [ora feorante ae [Caren crtas [Maras [" caternm [Cores [ora [i crrane Similarly. we can show that Ex-y = 4" Therefore Ey = E. + Ey. We are tempted to conclude thet Evay = E, ~ Ey in goneral. Let us see. ) =f tars f fo [ pare Ofer Eye [ coarse f (2a =a 0 fia fo ‘Therefore. in general Exsy # Ex + Ey 21-8 Exey (-1Pae= e,= [ore . Exes 1 faa a f® : Penk [Oct emttoat ronan az, | [+ con Quer +2] ce _ ia o co “SU asf cou dant + 20a] Flh+ol a 21-4 This problem is identical to Example 2.26. except that 21 wa. In this case. the third integral in Py (see p. 19 is not zero. This integral is given by ra hye Jim 2am f cas (wit + 61).008 (ait + Ba)dh 1m Jo a a = ttm S12 if cost; ~0)dt= f — cos(2ust +1 +62) dt oa “ra = fim, LEE Ir costes ~ tn)} + 0-= CrCacostAs — He) 1Therefore 7,4 Pym G+ F 4 CrCocosies - 82) 21-5 2 anil (ytd = 64/7 (@) Pe =if (Pur = 64/7 a a). ; Peep L fra Sign change of a signal doesnot afer its power, Multiplication of a signal by a constant « increases the power by a factor « 21-6 () Pa 2 4(64/7) = 256/7 (0) Pe 5 feta oer @ if (eat a w mek [Eye 21-7 1 f7? . a7? nella meet pre gin [ane dee jim $f de" a ra 7/2 fem tam “The integrais of the cross-product terms (when & 9 r) are finite because the integrands ave periodic signals Ihnade up of sinusoids). These terms. when divided by T — oo. vied zero. The remaining serms (k =r) yield re . ting f Sister =D Dal 1/2 fem ioe P, 2.1.8 (a) Power of a sinusoid of amplitude C is C2/2 Eq. (2.6a)| regardless of its frequency (w # 0) and phase Therefore. in this case P = (10)?/2 = 50. (b) Power of a sum of sinusoids is equal to the sum of the powers of the sinusoids Bq. (2.6b)]. Therefore. in this case P = 292 4 OF? = 178. (c) (10 + 2 sin 31) cos 10" = 100s 101 = sin 13¢ — sin BY Hence from Eq. (2.60) P= S95 + 4+ 4 = 51 (d) 10cos 54 cos 10t = 5(cos 51 + cos 18¢. Hence from Eq. (2.60) P sy 4 Yt = 05. (6) 10sin 5 coe 101 = sin 151 ~ sn Bt, Hence from Eq, (260) P= Se + EF = 28 (f) 0/9" coswo! = 9 [eter + 00-07]. Using the result in Prob. 2.1-7. we obtain P = (1/4) + (1/4) = 1/2 2.2-1 For areala For imaginary a. let a = jr, Then rina [emer tn [a P,J lero) Fig. $2.8-2 Clearly. if¢ is veal. ¢~*" is neither energy not power signal. However if ais imaginary, it is @ power signal with power 1 2.3-1 salt) = alt 1) rast 1). galt) = alt= 1 talt +2). galt) = alt ~ 0.5) + nit = 05) ‘The signal gs(1) can be obtained by (i) delaying g(t) by 1 second (replace t with t — 1). (i) then time-expanding by n factor 2 {replace f with 1.'2). (ii) then multiply with 1.5. Thus gs(t) = 1-59(3 ~ 1) 2.92 All the signals are shown in Fig. $2.3-2, 2.3-3 All the signals are shown in Fig. $2.3-3 AK gee) D fp 9G) 4 7] 9c) @ eo 6 ¢ se re 4k ge) 4p gee a jhe. °. 3 - ° Fig. $2.33 28-4 gaye [bana [Pina e, Baca = [it-opar= [7 steer = Bs taone [wrote fo etie= te Baas= [montanes [Peer lwne [ we nrare! [donee bue Baran [ptietane [doa ast of tomeee ‘24-1 Using the fact that a(7)A(r) = 4(0)A(7). we have (a) 0 (b) Fu) Ce) EMA) (4) $A 1) KD gable +3) (A) ko(w) (use LY ‘Hopital’s rule) = 0, Thus. an impulse A(t - r) is located st 2.4+2 In these problems remember that impulse (7) is located at ‘and 0 on, (a) The impulse is located at + =f and g(r) at 7 =f is alt). Therefore 324-3 25-1 2.52 25-3 2d [i sore ner = a0 (0) The impulse Nr) iat += 0 and glf = 7) at r= 0 sg). Therefore [i Moae-nar=ao Using similar arguments. we obtain (1 (40 (ee? (YS (w) 9-1) (h) -0? Letting at = 7. we obtain (for a > 0) £ ott stan at = im ot2)e(a}de = Lo(0) Similarly for a < 0, we show that this integral is -16(0). Therefore 1 0(t)6(t) dt [eomanan a= Therefore Hat) = Lan) Trivial Take the derivative of je'? with respect to « and equate it to zero [fovnta? ffsaeos = 0.5 over (0 <1 < 1). and zero outside this intervai, Also Ey (a) tn this case Ey = fi dt = 1 and (b) Thus. off) ¥ 0.5r(¢). and the error o(t) = and Ee (the energy of the error) are = [eon frees we gem ['e-osraeim lo o lo ‘The error (1 = 0.5) is orthogonal to r(1) because [ (-0.8y(1)dr = 0 Note that Ey = «Ey + Ee. To explain these results in terms of vector concepts we observe from Fig. 21% ar hence vector e is orthogonal to the component x. Because of this orthogonality. the length-square of fg energy of (1) is equal to the sum of the square of the lengths of ex and ¢ [auin of the energies of -7() and (| In this case Ey = J o2(Ndt = fy at = 1/3. and Ef sone afirmers “Thus. 2(t) © LS9(t). and the error e(t) = x(t) ~ 1.59(1) = 1 ~ 1.5 over (0 < 1 < 1), and zero outside this ‘Also E, (the eneigy of the error) is Ey = fi(1 — 1.80)? dt = 1/4 (a) In this case Ey = [2 sin? 2ntat = 05, and tin Qetdt = Afr altya(t)dt = (b) Thus. olf) = —(1/a)2(0). and the error e(t) = t+ (I/x)sin 2xt over (0 ¢ tS 1). and zero outside this interval. Also E, and Ee (the energy of the error) are 425-5 ft —(1/n)sin 2et}? at =} p -f sina [Pama and & ‘The error jt + (1/x) sin 2n/] is orthogonal to (t) because [ss ates cinan aenai=o Note that Ey = (Ey + Be. To explain these results in terms of vector concepts we observe from Fig, 2.15 that aoe, estar e is orthogonal to the component 7 i> “3 2 t Fig. $2.8-1 ‘The power of a(t) is Moreover. from Parseval's theorem [Eq. (2.90)] nad S$ () 45 (SF (b) If the V-term Fourier series is denoted by (1), thenFor N = 1, Pe = 0.1111: for N = 2. Pr = 0.19923, For N = 3, Py = 0.19837. which is greater than 0.198. Thus. N = 3. 2.8-2 Here To = 2n. so that wo = 2n/2n = 1. and att) = 204 So an coe tb Fig. $2.6-2 The power of att is aokf wae e Moreover. from Parseval’s theorem (Eq. (2.90)} 2 a ed LG (b) If the N-term Fourier series is denoted by r(t). then nnencut Soden ners The power Py is required to be 0.90 x $2 = 0.2, Therefore For N= 1. Pe = 2; for N = 2. Pp = 25. for N = 5, Pr = 2.927, which is less than 0.3n7, For N= 6. Ps = 2.9625, which is greater than 0.3n?, Thus. N = 6. 28-8 Recall that 1 pen won ef ae (1a) 1p ne (1) cos nwt dt {1b} To ria” > pan ; - by = Bf. wom nwol (le)2d Recall also that cos nent is an even function and sin nupt is an odd function oft. If g(t) is an even function of Pee at tes mast Ie also an even function and a(#)sin nunt is an odd function oft. Therefore (see hint) tot none f attde (29) ret one [arcs naott (2b) ino ro) imitans if 9) is an odd function off then (toe moot isan odd function off and g(t)in not it an even function of "Therefore fomeeeD (3a) 1 pm ere [ a(t) sin nwotdt (ab) mh, Observe that, because of symmetry, the integration required to compute the coefficients need be performed over only half the period (a) To = 4. a0 = #f = §. Because of even symmetry, all sine terms are zero. a(t) = 90+ San cos (Fe) ‘20 = 0 (by inspection) Therefore, the Fourier series for a(t) is: dy nt 1 tnt 1 Set Lag Tat (us - Sen t+ Jeon - gon Ft +) alt) Here hy = 0. and we allow Cv to take negative values. Figure S2.8-4a shows the plot of Cx (b) To = 107, wo = # = $. Because of even symmetry. all the sine terms are zero aim sos Domes (51) + pain (2) 1 20 (by inspection) oe kf G)e- FQ GL.) 2 fT aa(m ; ' bas zl (B) dt =0 (integrand is an odd function of t) ete hy = 0, and we allow Cy to take negative values. Note that Cy = an for n = 0, 1.2.3, Figure S2.8-4b shows the plot of Cn (e) To = 28. wo = 1 a(t) = 40+ 3 a0 cost + by sinnt with ag = 0.5 (by inspection) 20 a an=i fo Leosnrar mak [” Lsinntat =-2 7), 2 a), and \inai +2 gnats deiner +) nosed fn(-eg)ofem(erd) be 73 8> ce) Pig. $2.84 ‘The reason for vanishing of the cosines terms is that when 0.6 (the de component) is subtracted from 9(t). the emaining function has odd symmetry. Hence, the Fourier series would contain de and sine terms only: Figure $2.8-4c shows the plot of Cy and @ (a) To = x. wo = 2 and g(t) = 4. = 0 (by inspection) n= 0 (n>0) — because of odd symmetry 1 on aity= Sysinar~ Eoinat~ shy sme sin8t + aFigure $2.8-4d shows the plot of Cy and 4p. (0) To = 3. wo = 28/3. Therefore Co = } and 3 arn? Qen den Onn oa (32 cos 252 - sin 252 Coe — 2cos - sin anc = tan’ a Tent [\= y ~Beee— Z sim TT | and Oo tan | coe aa Be sin ee I (f) To = 6. 2 = 7/3, uo = 0.5 (by inspection), Even symmetry: by = 0. ap’ ne nnd [snc %a 23 [foe Fare [enon geal Observe that even harmonics vanish. The reason is that if the de (0.8) is subtracted from o(?). the resulting function has half-wave symmetry. (See Prob. 2 8-6). Figure $2.8-4f shows the plot of Cn An even function geit) and an odd function gel) have the property that Gelt) = gel-t) and Every signal g(7) ean be expressed as a sum of even and odd components because alt) = Flats) + af—0)] + $ lal) - of From the definitions in Eq. (1). it can be seen that the first component on the right-hand side is an even funetion. while the second component is odd. This is readily seen from the fact that replacing by ~1 in the first component yields the same function. The same maneuver in the second component yields the negative of that component. To find the odd and the even components of g(t) = u(t). we have g(t) = gelt) + aoit) where {from Eq. (1)] gelt) = $fu() + u(-1)] = 2 and galt) = 9 fut) —w(-0) = Zoantey 32.850, u(t), we have ‘The even and odd components of the signal w(t) are shown in Similarly, to find the odd and the even components of g(t) alt) = art) + welt) where aelt) =} [ott ult) + o*u(—1)] and 10te) Fig. $2.8-5 golt) = 3 fertu(n) = e*4u(—)] ‘The even and odd components of the signal ¢~*n(t) are shown in Fig. $2.8-5b, For gif) = 07". we have wt = oelt) + 9alt) where galt) =} [Ph + 0H] = e081 and galt) fe = 0-H] = jain 2.8:6 (a) For half wave symmetry a= -0(1= B) and and a 2 ft 2 sr an F | ateycosmoot dt = zf atoycosnantat» f a(t) cos nwot dt To Jo Te Jo ria Let + = 1 - To/2 in the second integral. This gives nb [[ aesnoerars ("0 (2+ B) caren (r+ F) «| on ns 2b [[Paomnnee [recent -4 [ f ater neat] Ina similar way we can show that 4? " if a(t)sin naot d (b) (3) To = 8. 0 = $. 0 = 0 (by inspection). Half wave symmetry. Hence uTherefore tr (FF -1) = 1,8.9,13. gfe (R41) n23.7.10,15, Similarly and a(t) = > 5.608 Et + bn sin nein (ii) To = 25. wo = 1. 00 = 0 (by inspection). Half wave symmetry. Hence at) = 3 faq cos nt + bn innit onl aonb fe cosmtat 2 [ee 7 22 [ gain em evan] ( - stn] and 20 2 22 [ee tondam-neanl] 2m igen) = 2A = Grom! Deyo 2.061 (a): To = 4.0 = 2/2. Also Do = 0 {by inspection). (b) To = 10%. wp = 24/108 = 1/5 a(t) = > Doe’S'. where odd) odd) 2 Inj2ce) (a) (2) wh Fig. $2.9-1 fe) alt) = Do nei", where, by inspection Dy = 0.5 a } r>0 spertare gh. other (Daim gles and wn {h (A) To = *. wo = Zand Dn =0 ait) = So Darl. where man (Lan ott)(e) To= 3.00 a= 3 Dee, whee pail Therefore : ae aha (Em (f) To = 6. 20 = 7/3 Do = 05 2.02 a(t) = Scout + sin (5¢ ~ For a compact trigonometric form. all terms must have cosine form and amplitudes must be positive For this reason, we rewrite g(f) as air) = Seon + 0ca (s¢ - ZZ) +2008 (8-5-7) sear + coe (t= 2E) +2008 (81 Figure $2.9-2a shows amplitude and phase spectra. (By inspection of the trigonometric spectra in Fig. $2.9-2, we plot the exponential spectra s shown in Fig. $2.9-9b. By inspection of exponential spectra in Fig, 2.022, we obtain ane deraened [ose erate + [ost gests] + (fre) (te Set (Got) ome (em 142.9-3 (a) g(t) = 2+ 2cos(2t ~ x) + cos(3t — = 2 2cos2t + sin3t (b) The exponential spectra are shown in Fig, $2.9-3. (c) By inspection of exponential spectra toy [eg eat 4 Ff 4g IH] = 24208 (2t — w) + 008 (34 — 5) (a) Observe that the two expressions (trigonometric and exponential Fourier series) are equivalent Fig, $2.93 2.94 1p ptea nn Ds Te if J(t)cos nuot dt — f f(t) sin nwot dt Tera 2 If git) is even, the second term on the right-hand side is zero because its integrand is an odd function of 1 Hence, Da is real. In contrast, if g(t) is odd. the first term on the right-hand side is zero because its integrand is an odd function of ¢. Hence. Dx is imaginaryChapter 3 El oa) =f anermat f fol) is an even function of ¢.o(t)sin.st ip an odd function of t, and the second integral vanishes Moreovs", viijcoeur is an even function off. and the frst integral is twice the integral over the interval Oto oe. Thus when 9(/) is even 3.161 = g(t) cost dt ~ if g(t) sinat dt Gu)= f a(t) cost dt a) Similar argument shows that when g(t) is odd Gl) = -a f o(t)sinwtdt (2) If gl?) is also real (in addition to being even), the integral (1) is real. Moreover from (1) a =f stoeaatt = Ou) lo Hleuce Gis) is real and even function of ». Similar arguments can be used to prove the rest of the properties 31-2 ” Gye & Male)! f Gly“ do = Ef, Gla) de L [fh toteroaersinees f tant tot] since 'G\x) is an even funetion and 0,(w) is an odd function of w. the integrand in the second integral is Since 1C1-)) i an oad therefore vanishes. Moreover the integrand in the first integral is an even function of and therefore (w)] dur a= 2 [ ratericotn Therefore [G(w)| = 1/VaT Fe and @,(w) = —tan"*(). Hence 1 a L aE [ur = tan | 3.8 etey= J alt)en™ at Therefore : ote [econ and ol f a (ned 16B.1-4 (a) ) T T Ge f[ tte at = f ° lo 3.5 (a) * 7 m7) 2 9 () lcotrr + wrsinwt - 1] ‘This result could also be derived by observing thet g(t) is an even function. Therefore from the result in Prob. 311 Ge = 2f tor atdt = 2aleoswr + ursina — I) 3.1-6 (2) I? _ (uBi? ~ 2)sin sot + 2a cos! —eewe Siw) Go? =2Z o =r fe + Fig. $31.6 (b) The derivation can be simplified by observing that G(-).can be expressed as a sum of two gare functions Gilw) and Gala) 98 shown in Fig. $3.1-6. Therefore sine d [overscan ao= Ef f ease f otda) = BBE 3.1-7 (a) ane [conse de B 12 = a litconw tana la = az (7) (b) Gaye ae = se iC Gieconande +3 Glu) sinwt «| ecause Glu) is even fusetion, the second integral on the righthand side vanishes. Also the integrand of the first term is an even function. Therefore Ww31-8 (a) emi den bf” otras -—1 mere inwolt - to) 7 - ae 1 = SED = Baineloote— to) &) oo=2[f sertans [seme] 1 yutl? 1 geal _, Lo coswot sa Lome . ects ay tect (E) 4p D038) ‘ rect ( e ) 7 Fo rt ZY = sT4 = me lo so a. a C 3 oF w * 4 sine C eum i Fig. $3.21 jous functions. The function in part (a) is a gate function centered at the 3.2+1 Figure S3.2-1 shows the plots of v origin and of width 2. The function in part (b) can be expressed as 0 (qgiyn)- This isa triangle pulse centered in and of width 100/3. The function in part (c) is gate function rect(§) delayed by 10. In other 'a gate pulse centered at t = 10 and of width 8 The function in part (d) is a sinc pulse centered tr the origin and the first zero occurring at * = x, that is at w = 5. The function in part (e) is a sinc pulse at the ori words it ne(Z) delayed by 107. For the sinc pulse sinc(), the first zero occurs at § =r. that is at w = Sx. Therefore the fonction is a sine pulse centered at w = 10 and its zeros spaced at intervals of 5 as shown in the fig S32-1e. The function in port (f) is a product of a gate pulse (centered at the origin) of width 10x and @ sine pulse (also centered at the origin) with zeros spaced at intervals of $x. This results in the sinc pulse truncated Deyond the interval 5x (|"{ 2 5x) as shown in Fig. £ 8.2-2 The function rect (f ~5) is centered at f = 5. has a width of unity. and its value over this interval is unity. Hence auye femam 2 oe _ gn98%) pit ity = sine (3) 7% 183.23 B24 3.25 3.41 e+ 2 ody = oh [einen — 20-7] 28 Joe Fire = Fpaltisiontl Observe that 1+ sgn(t) = 2u(t). Adding pairs 7 and 12 in Table 3.1 and then dividing by 2 yields the desired result. Observe that c08 (wo! + 4) Fourier transform of the above equation yields the desired result. (a) u(t) exe niu) 4 ~ is cy toy Application of duality property yields mbit) + 2 exe 2nu(-w) i" ma aaa) faved] meen Application of Eq. (3.28) yields Hyena Hien] ae Lyne Lime (b) cos wot => t16(u + wo) + Mw - x0) a[d(t +00) + A(t — wo)] => 2 cos (—wow) = 2m cos (wow) Atta Mt on om Bream) Aran eee-T) mo tenre te) sin wot => jr[A(w + wo) — H(w ~ 0)} jn IA(t + wo) — A(t = wn)] Qn sin(-wow) = —2n sin(ax dsin(t + 20) — ¢ N= i) (ow) Setting 20 = T yields 19at+ 7) = Ot = T) om isin Tw 3.3-2 Fig. (b) g:(t) = 9(-t) and Gs(0) = G(-w) = Fee ioe = Fig. (c) on) = off —1) + oul! = 1). Therefore Galw) = (Glu) + Gie)le™* = [Glw) + G(-w)e"" 2 (coo + wins ~ 1) Fig. (4) an(t) = g(t ~ 1) + alt +1) Galw) = Glw}e™ + G(-w)e* Fete ~ 2eoss} = sin? § = sine? (3) Fig. (e) ga(t) = o(t - 9) + alt + 4), and Gale) = Cla )e"™? + Gi(wer"? ort Fig. (f) asit) can be obtained in three steps: (1) time-expanding g(t) by a factor 2 (i) then delaying it by? ‘seconds. (ii) and multiplying it by 1.5 [we may interchange the sequence for steps (i) and (li). The first sten (time-expansion by a factor 2) yields t a) = Be(e = jdue™ — 1 (§) — 26020) = pile ~ Boe — 1) Second step of time delay of 2 secs. yields 4 pa “The thie stop of multiplying the resulting signal by 1.8 yields 14 = jaye! = 1)e j= ott) = 154 (532) ome G01 i200) cb? s)nce-T) ele-7 cost ult 7* (e-E) ut) 2 sintuleds Fig. $9.38 ayn (SEL) — ne ( wa ($) = 74() ( 132) — rane(sE) e407” 3.33 (a) 20Bde and Gt) = Tene (SE) fort? — 77) = 2jTsine (#) sin 2 ~So(F) (b) From Fig. $3.3-3b we verify that. a(t) = sin tu(t) + sin(t ~ #)u(e - *) Note that sin(t ~ x)u(t ~ =) is sin f(t) delayed by x. Now. sintu(t) ene GIA(w~ 1) ~ Ol + 1)] + ror and sin(t = )ult = #) em Flblw = 1) = Mes N+ i ‘Therefore 1 Glu) yee) {yl 1) — Ale +1))+ Recall that g(r)#(1 ~ 20) = a(roiM(r ~ 10), Therefore A(w + 1)(1 +677") = 0. and 1 Cle) peee™) (c) From Fig. S3.3-3¢ we verify that ait) = cost [rte = (1 §)] = costae) — cost But sinit - $) = cost. Therefore au) = costa) sin (= 2) u(' Gla) = FIMe +e D+ [ole = 1) — be + + Also because g(r)A(r — 70) = olru)4(7 ~ 70). Bet ye I = ak eI? = S joe £1) Therefore (4) u(t) = u(t - T)} = eMule) = elt) in(t) = ene enAl— T(t =P) Ve ier Jove jute” jove =e Gls) From time-shifting property a(t 7) — Gluyet™ Therefore g(t ET) + a(t — T) <2 Glu)o*T + Glue "7 = 2G(w) coswT We can use this result to derive transforms of signals in Fig, P3.3-4 (a) Heve off) is a gate pulse as shown in Fig. $3.3-4a 213.3-5 3.3-6 atte res ($) ae ait) to T= 9. The signal in Fg, P24 sof +3) +909) and g(t +3) + 9(t — 3) = Asinc(w) cos 30 (b) Here 9(t) is a triangular pulse shown in Fig. $3.3-4b. From the Table 3.1 (pair 19) y=) a9 sine (2) uo T= 9, The signal in Fig P-bit +3) + 9(0~ 9). and ait) + a(t 2)» Bain (2) cose t) 1 x w> t Fig. $3.34 Frequency-shifting property states that altietet exp Glo = wo) Therefore 1 a(t)sin cot = prieitiet™ + a(t ‘Time-shifting property states that 1 ) lO - +0) + GU alt £7) a Gaye? Therefore att =T) =~ oft —T) > Glu hel"? — Glue" = 2)G(w) sino T and Spelt =1)- alt 1) a Gle)ain To The signal in Fig, P3.3-5 is a(t +3) ~ ot ~ 3) where att)= nea (2) 2 ele) Therefore a(t +3) — 9(t — 3) <=> 2j[2sine(w) sin 3u} = 43 sinc(w) sin 30 Fig. (a) The signal 9) in this case isa triangle pulse (3) (Fig. $3.3-6) multiplied by cos 10¢ att) = & (x) cos 10" [Also from Table 3.1 (pair 19) (5) e=» + sine?(3#) From the modulation property (3.35), it follows that at 8 (g)o rem § Lone [22S] + sn resin} TThe Fourier transform in this case is a real function and we need only the amplitude spectrum in this case as shown in Fig, $3.3-6a. Fig. (b) The signal o(t) here is the same as the signal in Fig. (a) delayed by 2x. From time shifting properts: ne as in part (a) multiplied by ~~"), Therefore its Fourier transform is the 223.3-7 ine? { 22 = 10) a[xe+10]\ sae Gls {ain [pO] +e [ss ; |}. ‘The Fourier transform in this case is the samme as that in part (a) multiplied by ¢~/2*~. This multiplying fnctor represents a linear phase spectrum —2rw. Thus we have an amplitude spectrum (same as in part (a)) as well as 8 linear phase spectrum £Gl2) = ~2tw as shown in Fig. $3:-6b. the amplitude spectrum in thi ease as shown in Fig. $3.3-6b Notes In the above solution. we fist multiplied the triangle pulse A(z) by cos 101 and then delayed the result be 2x. This means the signal in Fig. (b) is exprersed as A(42*) cos 10(t ~ 27). We could have interchanged the operation in this particular case. that is. the triangle pulse A( 35) is first delayed ts 2e and then the result is multiplied by cos 10t. In this alternate procedure. the signal in Fig, (b) is expressed as A Jens 101. This interchange of operation is permissible here only because the sinusoid cos 10" executes integral number of cycles in the interval 2x. Because of this both the expressions are equivalent since cos 10(¢ ~ 2) = cos 10 Fig. (c) In this case the signal is identical to that in Fig. b. except that the basic pulse is rect (3k) instead of a triangle pulse A(z). Now 2 ret (sh) o- aesnctee) Lng re same argument a fr pat (9) bea Glu) = n{sine|x(w + 10)] + sinelr(w — 10)}}0- ) orm) ome (284) Also Saige) en vee (2) Therefore att) = Zaine(t)cosat (o) Also 23pana (2) Therefore a(t) = Ssinc?(t) cost 3.9:8 From the frequency convolution property. we obtain 90) = EGl) + Ge) ‘The width property of convolution states that if ¢(z) « ca(z) = y(z), then the width of u(x) is equal to the sum of the widths of ra(z) and a(x). Hence, the width of G(s) » Glo) is twice the width of G(w), Repeated ‘application of this argument shows that the bandwidth of ”(¢) is nB Hz (n times the bandwidth of 9(t)). 3.3-9 (8) 7 amt ge [ etaee 21 - A ain? (2 5 ow f 7 af dew 21 ~con at} (4) &) and Ge) = Taine (22) [077 = 0 77] = 2Teine (2 = Bent © S aay 47) -2H0) 460-7) The Fourier transform of this equation yields JT 9g iM? = 91h ~ coe wT} = ~4sin? (SE joG(w) = oT 2+ = ~2[1 ~ c06 w] = ~4sin? (2) Therefore oe) = Han (2 3.3-10 A basic demodulator is shown in Fig. $3.3-10a, The product of the modulated signal ¢(t)coswot with 2coswof yields a(t) coswot X 208 wot = 2a(t) cos? wot = 9(t)[1 + cos 2uot] = 9(t) + 9(t) cos 2u0t ‘The product contains the desired 9(f) (whose spectrum is centered at w = 0) and the unwanted signal 9(f) cos 2o! with spectium }|Gla-+2v0]+Glu—2u0!. which is centered at +20. The two spectra are nonoverlapping because 24gltjeosat glt> GAL@>D 29200. 2germeT we Fig. S341 UW < ao (See Fig, $3.3-10b). We can suppress the unwanted signal by passing the product through a lowpass filter as shown in Fig $3.3-108. Gi(e) = sine( gop) nd Gao) = Figure $3.4-1 shows G;(~). Ga(w). Hi(~) and Ha(w). Now Gi(w)Ha(w) Ga(u)Halw) ‘The spectra Yi(w) and Y2(w) are also shown in Fig. $3.4-1. Because v(t) = 1n(t)va(t), the frequency convolution property vields ¥(a) = Yiu) + Ya(w). From the width property of convolution, it follows that the bandwidth of Ye) is the sum of bandwidths of Yi(w) and Ya(w). Because the bandwidths of Yi(w) and Ya(w) are 10 kHz. 5 kHz. respectively. the bandwidth of ¥(w) is 15 kHz. Hw) Using pair 22 (Table 3.1) and time-shifting property, we get Leto 8 nt) = Ghee ) Vank ‘This is noncausal, Hence the fier is unrealizable. Also ™ [an iH (I) ~ bw ge 23.5-2 3.5-3 3.54 Figure $3.5-1 Hence the filter is noncausal and therefore unrealizable. Since ht) ia a Gaussian fonction delayed by fait looks Fete ae ase ndjecent figure. Choosing to = 3V2K, h(0) = ¢-*® m= 0.011 of 1.1% ofits peak value, Hence toc VHF is a reesonable choice to make the fier approximately realizable 2x10 soto ore 108 From pair 3, Table 3.1 and timeshifting property, we get Hw)= h(t) = tal ‘The impulse response is noncausal. and the filter is unrealizable. Figure $3.5-2 ‘The exponential delays 10 1.8% at 4 times constants. Hence fo = 4/a = 4 x 107° = 40us is a reasonable choice to make this filter approximately realizable. From the results in Example 3.16 et aeckeit HO = oes Re 7 Also HO) = 1. Hence if w1 is the frequency where the amplitude response drops to 0.95, then 108 Hol = Wool = Toe Moreover. the time delay is given by (see Example 3.16) = 0.95 ==> ws = 328.684 0-8 talw) = = t4(0) = 2 are If 02 is the frequency where the time delay drops to 0. 98% of its value at w = 0. then 10° tuber) = aa We select the amaller of wy and w2. that is w = 142,857, where both the specifications are satisfied, This yields 1a frequency of 22.7364 Hz. 0.98 x 1078 => wy = 142,887 ‘There is a typo in this example. The time delay tolerance should be 4% instead of 1% eee ea 900 centered at w = 10° represents the frequency range from 0.99 x 10° to 1.01 x 10°, Let tus consider the gains and the time delays at the band edges. From Example 3.16 1a * ae At the edges of the band 26S| = = «1077, x = = x 107% (10.90 « 30) = eer = 104 30-%, and UHCLOL 20 = eras = 990 * 10 ‘The gain vasiaton over the band is only 1.99%, Similarly, we find the time delays atthe band ease at qrwcttesro = oor and ta(1.01 10°) = qrorci esr = contr -phe time delay variation over the band is 4%. Hence, the transmission may be considered distortonless ‘The Signal te tranemitted with a gain and time delay at the center of the band, chat is 9 = = 10°, We also find TH (10°) = 0.01 and 4(10°) = 1-7. Hence, if g(t) is the input. the corresponding output 1s u(t) = 0.01 9(t ~ 1077) t4(0.99 x 10°) git) 6 to J £aletsT> ES-4-7) Fig. 83.6-1 3.61 2) ecstrertene V (a) = G(wprect (Fe x Gta) rect (G25) I~ ssn] This follows fom the fact that rf & 14.7 when + € 1. Moreover, Gla)rect (fp) = G(w) because Gl) i» bandlimited to B Hz. Hence V2) = G(u)e 9 = jkO() sin oT OE Moteoser. we can show that (see Prob. 3.35) Hott +7) = a(t T)] ot Ole) sino Hence Wt) = ft to) + Blalt ~ to 7)~ alt 19+ 7) Figure $3.6-1 shows o(f) and w(t) 40-2 Recall that the transfer function of a ideal time delay of 7 seconds is ¢"7*7. Henee, the trausfer function of the equalizer in Fig. P3.6-2 is Hag (w! = 9 + aye Et age A + age Ideally. we require the equalizer 10 have ealeMNanvea = saa fea amines ™ TEae Toa et mae PA ge AE te BE ot “The equalize in Fig. P26-2 approximates thi expression if we sleet ao = 1. ei = a. 02 Oe (aayte a3.761 Letting £ = Jy and consequently di = Syd 1a [~ tng VE Ben ging [P= ie Also from pair 22 (Table 3.1) Letting aw = Jy and consequently do = zbyde a Viet span |e a 3.7-2 Consider a signal ait)=sine(t) and (2) = Erect (32) ~entanate 2 [OB lea) # -mf eri 3.723 Recall that f Gale)“ dv and f ault)er dt = Gal) aa Therefore £ aa(aa(t) dt = af. alt) Ue Galw)e™" «| dat ef. Galw) ie ane « tom gf o(-wiGate)te Interchanging the roles of gs(t) and g(t) in the above development, we can show that [sonnet ef. Gx(o)Ga(-w) do 3.74 In the generalized Parseval's theorem in Prob. 37.3, if we identity ox(t) = sine (21 — mz) and x(t) = sine (2"Bt ~ nm), then cute) = gree (; ap) and Ga [lL ntomera= in [lm Cea Bur rect (3p) = 1 for lel < 2xB. and is 0 otherwise. Hence Therefore 28[-ntoncnne wi [roeen (Sy En In evaluating the integral, we used the fact that ¢#/2** = 1 when k is an integer. 3.7-5 Application of duality property (Ea, (3.24)] to pair 3 (Table 3.1) yields ‘The signal energy is given by ‘The energy contained within the band (0 to W’ ) is mace Bem] If Ew = 0.99E,. then 00) ae Wm 288 aye = Oe B76 14 (21) ane A(z). then the output (2) = Ale)H (w). where H (2) is the lowpass filter transfer function (Fig $3 7-6). Becanse this filter band Af — 0. we may express it as an impulse function of area 4n/. Thus, Hilo) = [ar flale) and ¥(w) = (4nAw)AS]O(w) = [4eA(O)A SIM) Here we used the property 9(7)8(r) = 9(0)A(z) {Eq, (1.23a)]. This yields wit) = 2AQ)as Next. because q2(1) <=> A(w). we have aye fo ane" dt so that a0o)= f~ gt)dt = Ey Hence. 9(t) = 2E 4. a Aiag $12 eres Fig. 83.7.6 B.8-1 Let g(t) = m(t) + aa(t). Then [ar(t) + eatt)jion(t +7) + 92(t + 7D} dt = Ry (7) + Roalt) + Rorea(7) + Roaes (7) where 1 lim = ru(t +r) at T Hie Reylt) = 29If we let gi(t) = Cr cos(wit + 41) and ga(t) = C2cos(wat + 42). then 7 Rant)” tm ef CrCacostwrt + 01) cos(wat + war + Aa) dt 7A According to the argument used in Example 2.2, the integral on the right-hand sie ie zero. Hence Reval?) = 0. Using the same argument. we have Ryag, (7) = 0. Therefore alr) Rand +Rnin) = Seoenrs Ger ‘This result can be extended to a sum of any number of sinusoids as long as the frequency of each sinusoid is distinct. hence. if = mot a(t) = 2 Cu cosinsot +04) then Relt) =F cosnwor Moreover. for go(t) = Coy Rgp(7) = C8, and fe Regoos(7) = nim $f CoCr coslant unr +6,)dt = 0 2 Thus. we can generalize the resuit as follows. If 8+ 5 Bcomaor alt) = Cot S_ Cn cos(nwat + On) then Rel) and . Splee) = BOBO) # FY Caio ~ moo) + (9 + nwo)] Fignre $3.8-22 shows the waveforms r(t) and 7(f — t) for 7 < Te/2. Let T = NTe. On the average. there are 1/2 pulses in the waveform of duration T. The area under the product n(t)x(t — 1) is N/2 times (3 ~ 7) 9s Shown in Fig. S3.8-2b. Therefore mania in} [arene 7A For J} < {rl < To. there is no overlap between pulses, and Re(r) = 0. For Ta < |r] < 7. pulses again overlap. BoP ere coer hal pulses overlap. Hence. Rea(r) repeats every Ty seconds, but cnly with hal she paitude, as shown in Fig. S382. We can express Ro(r) es sum of two components. fs Sow ae ee Tee) = Ralt) + Ra(*). The PSD is the sum of the Fourier transforms of Ra(r) and Ral) Hence wy a where S2(w) is the Fourier transform ‘of the periodic triangle function, shown in Fig. $3.8-2d. We find the ‘exponential Fourier series for this periodic signal to be Seu) = B sine? (2) + S414) Ralt)= So Da we shsine(4z). Hence, according to Eq. (3-41) Using Eq. (2.80). we find Dn sw=F sine? (SE) ao- ran) ore 30© (a oc ae Aten T Ber (b) * Ne, % % %e Me “2% -35 -h =I a bv 3 T . 3 he 3 Eo 24 o> (e? ¢ RID) 3/8 “bis Tee to uel ated St -3T -k & oT B ly a2 — S (4) Sl) -oF vat s A Te ws % Fig. 93.8-2 Therefore sats) B sie? (22) +3 So sinc? (SE) sto-mn) oe 8.88 M2) = shy and |Htw)!? = sy. oie1 aye) " =7 and Wo-2f aeei ws Foe b Mead [Meee aj, wel 7 8.8-4 The ideal differentiator transfer function jw. Hence, the transfer function of the entire system is, 1 ie oe (Ga)w eta a aw Fij= 2 [we (3) oe we Fiat (me (2) fae 32Chapter 4 ES 4.2-1 (i) For m(t) = cos 10007 Poen set) = 1 (t) c08 10, 000t = cos 1000t cos 10, 000¢ 1 $leq9.90004 + cos 11,0001 we vsB, Gi) For mit) = 2co0s 10001 + cos 2000 “eon se(t) = (1) cos 10, 000t = {2 eos 10001 + cos 20001] cos 10.0001 29000" + cos 11.0001 + 3 feos 80001 + cos 12, 0001] = {coe90001 + # cos 80001] + [cos 11.0008 + 4 cos 12.000¢] ————_—— ns tee Lait) For m1) = cos 10001 cos 30001 Ente se(t) = m(t)c0s 10.000 = 3 {eos 2000¢ + eos 40001] cos 10, 000t 2 1 cos 8000! + cos 12.0001! + $ {cos 60001 + cos 14,0004] = sata osnn + 2 0 + cor 14.08 te use ‘This information is summarized in a table below. Figure S4.2-1 shows various spectra. Modujated signal sheehum Tapa 18 =e “6K Fig. $4.2-1 33Cb) a @) a na Mey 4200-~ mao (6) 2 (> 2 wh use 168 4 ise vse Fig. $4.28 tase | Baseband frequency | DSB frequency | LSB frequency | USB frequency 7 1000 ‘T3000 and 17,000] 9000 | ~—-1n.000'— | ir 7000 ‘3000 and 11,000 | "3000 T__11,000 2000 000 and 17.000] 8000 | 12.00] ii 2000 ‘3000 and 17.0007 8000. -| 12000 —*| 000 000 and 14.000 6000 | 14.000 4.2-2 The relevant plots are shown in Fig. $4.2-2 4.2-3 The relevant plots are shown in Fig. $4.2-3. 4.2-4 (a) The signal at point b is dal) = tn{t) cos wet = nt [Seonue + Leoesal 34426 ‘The term Jin(t)coswet is the desired modulated signal, whose spectrum is centered at du. The remaining term Im(f}cosduct is the unwanted term, which represents the modulated signa with carrier frequency 3c tin spectrum centered at 43e a8 shown in Fig, $4.2-4. The bandpass fiter centered at we allows to pass the desired term 3ru(t) cos.vet, but suppresses the unwanted term vn(t) eos Suet. Hence. this system works as desived with the output 310(t) cose. (b) Figure $4.2-4 shows the spectra at points b and (c) The minimum usable value of we is 2rB in ‘order to avoid spectral folding at de. (a) (theos? wet = "AO + con 2st] eae = Jit) + fm) cosduet ‘The signal at point b consists ofthe baseband signal rn(?) and a modulated signal f(t os uct, which has 9 carrier frequency 2u. not the desired value we. Both the components will be suppressed by the filter. whose ‘center center frequency is we. Hence. this system will not do the desired job. {e) The reader may verify that the identity for cosnuet contains a term coset when mis odd. This is nor true when 7 is even. Hence. the system works for a cartier cos” wet only when n is odd ! a @ Fig S424 We use the ving modulator shown in Fig. 4.6 with the carrier frequency fe = 100 kia (< = 200% x 10%), and the output bandpass filter centered at fe = 300 kHz. The output v,(t) is found in Eq. (4.7b) as wud trent = neat Emin er] ‘The output bandpass filter suppresses all the terms except the one centered at 300 kHz (corresponding to the cartier 32<1). Hence. the filter output is 4, i u(t) = mt) cos Bet ‘This is the desired output kin(t) coset with k= ~4/3n. “The resistance of each diode is + ohms while conducting. and oo when off. When the carrier A cos. is positive the diodes conduct (during the entire positive balf cycle). and when the carrier is negative the diodes art open {during the entite negative half cycle). Thus, during the positive half cycle. the voltage 5; 0(") appeais across cach of the resistors Tt, During the negative half cycle. the output voltage is zero. Therefore, the diodes act ss gate inthe eirenit that is basically a voltage divider with a gain 217/(R +r). The output is therefore 2R Foon) 1 fare. Hence. from Eq, (2.75) colt The period of w(t) is Toped? fou Love as Besa] 3 The output o(t) is cat) = 2 wept) = Bit [b+ 2 (con aut Leon du + Foor et +-~)] Teer 7 (n) If we pass the output cat) through a bandband filter (centered a int) coe nace! forall» 1s Jeaving only the modulated term sfx ‘as a modulator (b) The same circuit can be used as a demodulator if we use 6 baseps is o(t) = mu(?) coset and the output is 35K 20% “BK aw 20K BBK Ae me) cos 200006 cos teecet i> oe Fig. $4.2-8 bande. 4.2:8 (a) Fig, $4.2-8 shows the signals at points (b) From the spectrurn at point c. it is clear 5000 to 38.000 rad/s.) (c) Fig. $4.2-8 shows the receiver to recover 11 that the channel bandwidth must be at least 30.000 rad/s ({rom (t) and m3/t) from the received modulated signal. 4.2-9 (a) $4.2-9 shows the output signal spectrum ¥(#) (b) Observe that ¥(«) is the same as AJ («) with the frequency spectrum inverted. that is. ‘the high frequencies: 0 ovis to lower frequencies and vice versa. ‘Thus, the scrambler in Fig, P4.2-9 inverts the frequency spectrum 3642-10 say emapeesaznotivt) = Eh cone at fo soon 431 Fig. $4.3-2 ‘To get back the original spectrum A/(.c). we need to invert the spectrum Y (2) once again. This can he done by passing the scrambled signal »(?) through the same scrambler. We use the ring modulator shown in Fig. 4.6, except that the input is m(t)cos(2x)10% instead of 1(t), The cartier frequency is 200 kH2 {ure = (4007)10%). and rhe output bandpass filter is centered at 400 kHz, The output (1) is found in Eq, (4.76) as con s400e 10% + Bon satorn0% =] The product of the terms (-1/3)cos 8(400n)10% and (4/x)m(t)cos(2n)10° yields the desited erm widinnt) cos (800x)10"Y. whose spectrum is centered at 400 KHz. It alone passes through the bandpass fiter (centered at 400 kHz). All the other terms are suppressed. The desired output is 2 y(t) = u(t) cos (8007)10"¢ golt) = [A + m(t)] cost. Hence. ant) = [A + mit)]c08? wet 5 dia + m(o)+ HA + m(e)] conver “The first term is a lowpass signal because its spectrum is centered at w= 0. The lowpass filter allows this tenn to pose, but suppresses the second term, whose spectrum is centered at £2uc. Hence the output of the lowpass filter is u(t) = A+m(t) When this signal is passed through a de block, the dc terin A is suppressed yielding the output m(#), This shows that the system can demodulate AM signal regardless of the value of A. ‘This is a synchronous or colivent demodulation, 374.32 (a) = A=20 (b) 2 A=10 (©) = AaS «@) AO ‘This means that j1 = 20 represents the DSB-SC case. Figure $4.3-2 shows various waveforms 48:3 (a) According to Eq. (4.108), the carrier amplitude is A = mp/ = 10/08 = 128. The carrier power & P, = A?/2 = 78.125. mt) am % Fig. $4.3-3 Jf amplitude values every quarter cycle. the power of (b) The sideband power is 17(1)/2. Because of symmetry of fa quarter cycle only. Over a quarter cycle mn(t) can mut} may be computed by averaging the signal energy over Le iepresented as 1(t) = 401/To (see Fig. $4.3-3), Hence. ne mie Met met mf [ Yara saa ‘The sideband power is 2(t) mee p= OP = 1667 The efficiency is 16.67 7 = FETE STE * 100 = 19.60% nS Rt fis |m(t)]. The signal [A+ m(¢)]coswet is Jear that the envelope of the signal m(t) cos we ‘argument, it is cleat that t with m(t) replaced by A+ (t). Hence. using the previous ‘Now. if A+ m(t) > 0 for all f, then A+ m(t) = |A+:m(t)j. Therefore. the condition Jope detector is A+ m(t) > 0 for all 1. 4 From Fig. $4.3-4 it is el identical to m(t) cos its envelope is [A + (1) for demodulating AM signal using envel ma(tgs ces 27Tt[Armcesfees 2a [A- meey] c056,¢ Fig. $4.55 4.3: When an input to a DSB-Se generator is 1m (t) the corresponding output is mit) coswet: Clearly. if the inp is ‘ta m(}: the corresponding output will be [A'+ m(t)] coset. This is precisely the AM signal, Thus, by adding fe of value A to the baseband signal m(?), we can generate AM signal using » DSB-SC generetor ‘The converse is generally not true. However, we ean generate DSB-SC using AM generators if we use two identical AM generators in a balanced scheme shown in Fig. $4.3-5 to cancel out the carrier component 4.36 When an input to a DSB-SC demodulator is m(t) cosact. the corresponding output is 1m(t). Clesrs, if the inpatis [At ro (t)ic0s.aet. the corresponding output will A-+m(t). By blocking the de component A from this output, we can demodulate the AM signal using « DSB-SC demodulator. The converse, unfortunately: is not true. This is because. when an input to an AM demodulat the corresponding output is |1(t)j (the envelope of ru(t)]. Hence. unless m(t) > 0 for allt. demodulate DSB-SC signal using an AM demodulator. (1) coswet. not possible to 24 te -2A Fig. S4.8-7 4.3-7 Observe that w2(t) = A? for allt. Hence. the tine average of 2(t) is also A?. Thus wi) & 7 79 ‘The carier amplitude is A= up/u = mp = A. Hence Pe = A?/2. The total power is Pr = Pot P power efficiency is mje Aae Pe 2. The =f 10= n= Bx 100= AF x 100=05 The AM signal for = 1 i shown in Fig, $4.3-7. 4.8-8 ‘The signal ot point a is [A + m/(t)] cos wet. The signal at point b ix AP + Dammit) + m2) 4 coe Duet) (1) = [A + mi}? c08? wet = : The lowpass filter suppresses the term containing cos 2u Me =100| 100 a> De qo No 2a) Tc Bs la vse -Q00 wr Goo 00 CD Mc) Fy3D 100 30D Wp» =I3K -un4 “7h Io wiki ak Fuse oe FE vs ar 7 ak 77R [OD TR AB TIaK -R Oa Tk ISR (») Mus) Fas ik ae ig 1400 mee 100|O Yoo Gor 1400 [600 oer cee -4oo |O 00 $00 o> Bee Dina! ‘Tyeo-4e0 [> 400 00 Zam 1408 (600 o> (c>) Figures not tw seale - Fig. $4.5-1MCdHA,) + Mle d= Bi? ® Toot \oe0eT Fig. $4.5-3 Mia) w Seo # 0in SSB (o) Sw oo . Fig. 34.8-5 sa) Figure $4.5-3a shows the spectrum of u(t) and Fig. $4.5-3b shows the corresponding DSB-SC spectrum Init) cos 10,0807 & acon gfe 2c shows the corresponding LSB spectrum obteined by suppressing (Te USB spectrum. ae oad shows the corresponding USB spectrum obtained by suppressing 1 LSB spectrum (Fire 6s Sheay Fourier eansforms of uhe LSB and USB spectra from Table $1 (pair 18) and the frequency shifting property as 45-3 uan(1) = 1000 sine (10001) cos 90007" ean lt) = 1000sinc (1000x?) cos 11, 000% uted Because Mn(w) = —JM(«) sgn (u). the transfer function of « Hilbert transformer is H(w) = ~isen(o) If we apply: ma(t) at the input of the Hilbert transformer, Y(w), the spectrum of the output signal u(t) is ¥-(u) = An o)H (2) = [- JA (0) 9m 2] Jen (W)] = Ge) . mit). To show that the energies of om(t) and ven(t) are This shows that the Hilbert transform of 1a(t) is equal, we have En [moa é . (a (w)PP dw | wyrae= Eo [livnertas Ef mertinan res a24.55 The incoming SSB signal at the receiver is given by [Eq. (4.17b)] ¥esn (t) = m(t) coset + ma(t)sinwet Let the local eartier be cos [(ve + Ais)t + 6]. The prodeuct of the incoming signal and the local cartier is a(t) given by alt) = Frsalt) 08 [(we + Bw)t + 4) = 2[mn(t) conwet + malt) sinwet] cos [(we + Aw)t + 4) ‘The lowpass filter suppresses the sum frequency component centered at the frequency (2ue + Sw). and passes only the difference frequency component centered at the frequency Aw. Hence, the filter output co(t) is given by colt) = t(t)cos(Aw)t + 6) ~ rma(t)sin(Aw)t + 4) Observe that if both Aw: and 4 are zero. the output is given by eo(t) = m(t) as expected. If only * = 0, then the output is given by rolt) = m{t) cos(Aw)t ~ ma(t)sin(Aw)t This is an USB signal corresponding to a carrier frequency Aw as shown in Fig. $4.5-5b. This spectrum is the same as the spectrum A/(w) with each frequency component shifted by a frequency Aw. This changes the sound Of an ovdio signal slightly. For voice signals, the frequency shift within £20 Hz is considered tolerable. Most US systems. however.sesttict the shift to 42 Hz. (b) When only Sw = 0. the lowpass filter output is oft) = m(t)eosA — malt)sind We now show that this is a phase distortion, where each frequency component of A/(~) is shifted in phose by amount +. The Fourier transform of this equation yields = Mw) cos — Ma(w) sind But from Eq. (4.146) -jM(w) w>0 Mle) w <0 and ae w>o Mw)e*® wc It follows that the amplitude spectrum of ro(*) is AM(w). the same as that for m(t). But the phase of each component is shifted by 4. Phase distortion generally is not a serious problem with voice signals, because the human ear is somewhat insensitive to phase distortion. Such distortion may change the quality of speech. but the voice is still intelligible. In video signals and data transmission. however, phase distortion may be intolerable. 4.5-6 We showed in prob. 4.5-4 that the Hilbert transform of mra(t) is -m(t). Hence, if ma(t) [instead of m(t) is ‘applied at the input in Fig. 4.20. the USB output is ult) = men(f) cos wet = rn(i) sin wel nt) om (2 +2) at in (eat 4346-1 4.81 48-2 -4 -2 402 4 $ bia Fig. $4.61 Thus. if we apply ma(t) at the input of the Fig, 4.20. the USB output is an LSB signal corresponding to ("> The cairier also acquires a phase shift +/2. Similarly, we can show that if we apply ra(#) at the input of the Fig. 4.20, the LSB output would be an USB signal corresponding to mn(t) (with a carrier phase shifted by #/2) From Eq. (4.20) 1 Hole) = FIST) = Meo) fol s 2B Figure $4.6-18 shows H,(w ~ wc) and Hy(w +we). Figure $4.6-1b shows the reciprocal. which is Hel {A station can be heard at its allocated frequency 1500 kHz as well as at its image frequency. The two frequencies are 2/ir He apart. In the present case. fir = 488 kHz. Hence. the image frequency is 2 x 458 = 910 kHz apart ‘Therefore. the station will also be heard if the receiver is tuned to frequency 1800-910 = 590 kHz. The reascn for this is as follows, When the receiver is tuned to 590 kHz. the local oscillator frequency is fu = 890+ 455 = 1045 kia. Now this frequency fuo is multiplied with the incoming signal of frequency fe = 1500 kH¥z. The output sields the two modulated signals whose cartier frequencies are the sum and difference frequencies. which are 1500 1045 = 2545 kHz and 1500 ~ 1045 = 455 kHz, The sum carrier is suppressed, but the difference carrier passes through. and the station is received ‘The loca! oscillator generates frequencies in the range 1+8=9 MHz to 30+8=38 MHz. When the recciver setting is 10MH2, fio = 10+ 8 = 18 MHz, Now. if there is a station at 18 +8 = 26 MHz. it will beat (mix) with fio = 18 MHz to produce two signals centered at 26+ 18 = 44 MHz and at 26-18 = 8 MHz. The cum component is suppressed by the IF filter. but the difference component. which is centered at 8 MHz. passes through the IF filter 44 Hi (2-& y+, (+e,Chapter 5 ES init) #40 4h? | Ct t= tot 44 Fig. $5.1-1 5.11 In this case fe = 10 MHz. ny = 1 and ms = 8000. For FM : Af = kyinp/2n = 2m» 102/2n = 10° He. Also fo = 107. Hence, (f)mex = 107 + 16° = 10.1 MHz, and (Jdmn = 10" ~ 10° = 9.9 MHz, The carrier frequency increases linearly from 9.9 MHz to 10.1 MHz over a quarter (rising) cycle of duration a seconds. For the next @ seconds. when 1n(t) = 1. the cartier frequency remains at 10.1 MHz. Over the next quarter (the falling) cycle of uration a. the carrier frequency decreases linearly from 10.1 MHz to 9.9 MHz., and over the last quarter cycle, Shen m(t) = —1. the carrier frequency remains at 9.9 MHz. This cycles repeats periodically with the period 4a $0 x 8000/2 = 2.x 10° Hz. Also fe = 10". Hence. (f,)max = 107 +2 x 10° = 10.2 MHz ‘and (J.jmm = 107 ~ 2x 10 = 9.8 MHz. Figure S5.1-1b shows rn(t), We conclude that the frequency remains at 102 Mily over the (rising) quarter cycle, where ti(t) = 8000. For the next a seconds, rn(t) = 0. and the carrier frequency remains at 30 MHz. Over the next @ seconds, where mh(t) = ~8000. the carrier frequency remains at SSMS Over the last quarter eyelevi(t) = 0 again, and the carrier frequency remains at 10 MHz. This cycles repeats periodically with the period 4a seconds as shown in Fig. $§.1-1 51-2 In this case fe= 1 MHz. rp = 1 and an, = 2000. For FM: ‘Sf = bymp[2m = 20,0007/2x = 10° Hz, Also fe = 1 MHz. Hence, (f,)man = 10° + 10% = 1.01 MHz. and (Judmin = 108 - 10° 99 MHz. The cartier frequency rises linearly from 0.99 MHz to 1.01 MHz over the cycle (over the interval ~25-* <1 < 182), Then instantaneously, the carrier frequency falls vo 0.99 MHz and staris fising linearly to 10.0) MHz over the next eyele. The cycle repeats periodically with period 10" as shown in Fig. $5.14 For PM Here, because m(t) has jump discontinuities. we shall use a direct approsch. For convenience, we select the origin for m(t) as shown in Fig. $5.1-2. Over the interval > to AZ. we can express the message signal as s(t) = 2000". Hence. ratty cos [arose Em] nes 0s + E200] 00s [2r(10}81 + 1000nt] = eos [2x (10" + 500) ] [Av the discontinuity. the amount of jump is mg = 2. Hence. the phase discontinuity is kping = x. Therefore, the cartier frequency is constant throughout at 10° + $00 Hz, But at the points of discontinuities, there is @ 43 mt> . R000 on 7 7 = 3008 ad t= “Jor2M tof Gey OM 10-24we mit) t> Fig. $8.1 phase discontinuity of x radians as shown in Fig, $8.1-2b. In this ease, we must snaintain ky < Decause ‘here paw Giscontinuity of the amount 2. For kp > x, the phase discontinuity will be higher than 2 giving rise 1° in demodulation, 5.13 (a) gpailt) = cos [et + kprn()} = 10 ens[20, 0008 + kpr(t)} We are given that gpsi(t) = 10 cos (13.000) with kp = 1000. Clearly. m(t) = 3¢ over the interval |< (b)— grwilt) = Acos beef m(ayéa| = 10 cos [rome f mia) 4a] Thecetore by f/mfa)da = 2000 [torte = sone tierce ate fmfojdo m= 9 B.de1 In this case ky = 1000 and kp = 1. For sn(t) = 2 cos 1001 +18 €082000rt and x(t) = ~200 sin 100 ~ 36,0007 sin 20001 “Therefore nip = 20 and m= 36.000 + 200. Also the baseband signal bandwidth B= 2000x/2r = 1 kHe. For FM J = kying]2n = 10.000, and Bry = 2(A/ + B) = 2(20.000 + 1000) = 42 Kitz, Fo eee an x 16.000 + 198 Ha, and Bow = 2(Af + B) = 2(18.031.83 + 1000) = 38.06386 kits 5.2-2 yey) = 10 cos(wet + 0.1 sin 2000rt), Here, the baseband signal bandwidth B = 2000/2" = 1000 Hz. Also, wi (t) = we + 2007 cos 2000nt a Therefore. Aw = 200m and Af = 100 Hz and Br: = 2(Af + B) = 2(100 + 1000) = 2.2 kHz ‘ 8.2-3 Zeylt) = 5 cos(wet +20 sin 1000r¢ + 10 sin 20001). Here. the baseband signal bandwidth B = 2000” /2x = 1000 Hz. Also, ult) = we + 20,000" cos 1000x1 + 20,000 cos 2000n! 0 kHz and Bens = 2(f-+B) = 2(20,000-+1000) = 12 ‘Therefore. Aw = 20,0007-+20.000x = 40,000x and Af kHz. 5.2-4 The bascband signal bandwidth B = 3 x 1000 = 3000 Hz For FM: Af = Sgt = S21 = 15.951 kilz and Bru: = 2(A/ + B) = 37.631 kits, For PM: Af = “ze = 2858000 = 91.831 kHz and Br = 2A + B) = 66.662 kHz 5.28 The baseband signal bandwidth B = § x 1000 = 5000 Hz. For FM: Af = “Sct = M341 = 1 kHe and Bru = (Af + B) = 2(2 +5) = 14 kHz. 465.26 5.81 For PM: To find BPM. we observe from Fig. $5.1-2 that p(t) is ensentially a sequence of sinusoidal pulses of width T = 107" seconds and of frequency f. = 1 MHz. Such a pulse and its spectrum are depicted in Figs. 43.22c and d. respectively. The bandwidth of the pulse, as seen from Fig. 3.224, is 4n/T rad/s or 2/T Hz. Hence, Brn = 2 kHz. (a) For FM: Af = S422 = 200i00*1 = 100 kHz and the baseband signal bandwidth B = 250" = 1 kHe ‘Therefore Bry = (Af + B) = 202 kHz For PM: Af = S5%e = isp = 10 kHz and Bro = (Af + B) = 2(10+1) = 22 KH. (b) rn (t) = 2 sin 2000nf, and B = 2000x/2n = 1 kHz. Also mp = 2 and m, = 40007. For FM: Of = S482 = 22222? = 200 kits, and Bra = AS + B) = 2(200 + 1) = 402 KH For PM: (As + B) (e) mit) For FM Brn: = 2Af + B) = 2(200 +2) = 204 kHz For PM: Af = ‘22h = i0sgQ00s = 20 kHz and Bra = 2(AS + B) = 2(20 + 2) = 44 kil (4) Doubling the amplitude of (1) roughly doubles the bandwidth of both FM and PM. Doubling the frequency of m(ts jexpanding the spectrum AY () by 8 factor 2] has hardly any effect on the FM bandwidth. However. it oughly doubles the bandwidth of PM. indicating that PM spectrum is sensitive to the shape of the baseband speetium. FM spectruin is relatively insensitive to the nature of the spectrum 4 (~}. From pair 22(Table 3.1). we obtain oo? gee fe on!!, The spectrum Af (w) = vio is a Gaussian pulse. iehich decays rapidly, its 3 dB bandwidth is 1.178 tad/s=0.187 Hz. This is an extremely small bandwidtts compared 10 Af iAlso s(t) = —21e7/2, ‘The spectrum of in(t) is AI'(w) = jwM(w) = iV Ree". This spectrum also dears ‘apidly away from the origin. and its bandwidth ean also be assumed to be negligible compared to 4 For FM: Af = Me = M9244 = 3 kHz and Bry = 2Af = 2x3 = 6 kite = -2te~!/? equal to zero. This yields For PM: To find mp. we set the derivative of r( nen gate 1 tit) = 20°F 44 PO met ° 4 and my = vi( Jy) = 0.858. and Sf = Sse = Ms2ON8E = 5.432 Kll2 and Bpws = 2(AJ) = 218.432) = 6.864 kHz ‘The block diagram of the design is shown in Fig. $5.3-1. Oe BF. 1250 He @) 10-865 MHz Fig. $5.3-1 a7fer 3M HE $= 96" Afe 4 TEGGHE DF2625H2 Af 201 5.32 The block diagram of the design is shown in Fig. $5.3-2. B41 (a) peni(t) = Acos [vet + kym(t)] When this yeni?) is pasacd through on ideal FM demodulator, the output is kpri(t) This signal. when passed Through an idea! integrator. yields Hy). Hence, FM demodulator followed by an ideal integrator acts as s PM Gemodulator, However, if m(t) has'a discontinuity rn(t) = 00 at the point(s) of discontinuity. and the system will fi (6) grni(t) = A cos fea [ tarde] When this signal gru(t) i passed through an ideal PM demodulator, the output is ky J" rn(o}da. When this winel it bussed tIvough an ideal diferentiator, the output is kyrn(t). Hence. PM demedlator. followed by an sant ifferemator, acts as FM demodulator regardless of whether r(t) has jump discontinuities or not 5.4-2 Figure $5.4-2 shows the waveforms at points b. ¢. 4. and e. The figure is self explanatory 5.43 From Eq, (5.30). the Laplace transform of the phase error @.(t) is given by els) = Ox(s) THAR) For (1) = 13, O(n) = Band 2k Ol) = SRF ARAO] The steady-state phase error [Ea. (5 33) is Hence. the incoming signal cannot be tracked. If Hey= 22%, en Ol) =F we ay and 2 im 00) = Bagot = ki rape ay” As ence. the incoming signal can be tracked within a constant phase 2k/Aka radians. Now. if Stange 2 Hoy = AB then =F yep ] . and 2k in ntn= im STAR (ar ereh) ° In this case. the incoming signal can he tracked with zero phase error. 48Wu ALLL MTT + 8 WlChapter 6 Neen nn G.1-1 The bandwidths of gu(t) and ga(t) ate 100 kHz and 150 kHz, respectively. Therefore the Nyquist sampling rates for 41(t) is 200 kHz and for gat) is 300 kHz, Also 912(t) <=> grgi(w) * g1(w), and from the width property of convolution the bandwidth of 9:7(t) is twice the bandwidth of 4i(t) and that of ¢2%(t) is three times the bandwidth of g2(t) (se also Prob. 4.3-10). Similarly the bandwidth of 9:(t)ga(¢) is the sum of the bandwidth of 9:(t) and ga(t). Therefore the Nyquist rate fo a(t) is 400 kHz. for a2°(t) is 900 kHz. for 93(t)oa(") is 500 kH2. 61-2 (a) sinclOOnt) <=> 0.01rect (xi) “The bandwidth of this signal ie 100 x rad/s or $0 Hz, The Nyquist rate is 100 He (samples/sec) ) sinc?(100R¢) <= 0.01( a8) ‘The bandwidth of this signal is 200 x rad/s or 100 Hz, The Nyquist rate is 200 Itz (samples/sec) (ec) sinc (100xt) + sine (S0rt) <=# O.01rect 0.01 (qi) + 0.02rect (ee) ‘The bandwidth of the first term on the right-hand side is 50 Hz and the second term is 25 Hz Clearly the bandwidth of the composite signal is the higher of the two, thet is. 100 Hz, The Nyquist rate is 200 Hz ‘samples/sec) (s) sinc( 10071) + 3sinc?(60xt) <=> 0.01 reet( pz) + 3 MC aie) The bandwidth of rect(z3iz) is 80 Hz and that of 2( xZ5-) is 60 Hz. The bandwidth of the sum is the higher of the two. that is. 60 Hz, ‘The Nyquist sampling rate is 120 Hz ©) sinc(SOrt) exe 0.02 reet( ie) sine(100nt) «= 0.01 rect( sie) The two signals have bandwidths 25 Hz and $0 Hz respectively. The spectrum of the product of two signals is 1/2n times the convolution of their spectra. From width property of the convolution, the width of the convoluted signal is the sum of the widths of the signals convolved. Therefore. the bandwidth of sinc(S0x/)sine(100*t) is 25 +80 = 75 Hz. The Nyquist rate ie 150 Hz. 6.1-3. The pulse train is a periodic signal with fundamental frequency 2B He. Hence, w, = 2x(2B) = 4xB. The period is To = 1/2B. It is an even function of t. Hence, the Fourier series for the pulse train can be expressed as pri(t) = Co+ > Cn cos mast Using Eqs. (272). we obtain 1 pase o poe maton ed a ooncn=z f cosnust dt = To dines To Jspse sin ( alk yb Hence. 0) = apr.) at) + sin (22) a(t) os nest 3030 ry rd 20 Fay o 1 ton o> ed TD wy 03 Fig. 96.1-4 1-4. For g(t) = sinc®(Snt) (Fig, $6.1-48), the spectrum is G(w) = 0.2.(38;) (Fig. $6.1-4b). The bandwidth of this signal is 5 Ha (10% rad/s). Consequently, the Nyquist rate is 10 Hz, that is, we must sample the signal at @ rate ino less than 10 samples/s. The Nyquist interval is T = 1/2B = 0.1 second. Recall that the sampled signal spectrum consists of (1/T)G(u) = 44 O( 32) repeating periodically with a period equal to the sampling frequency f, Hz. We present this information in the following Table for three sampling tates: f= 5 Ha (undersampling). 10 Hz (Nyquist rate). and 20 Hz (oversampling). caning Raquney Te [mamplng intersal FOG) [comments | 5 Hz 0.2 ‘ (38) | Undersampling | Tore or TB (a) | Nyauist Rate 20 He 0.05 4B (ae) | Oversampling In the first case (undersampling). the sampling rate is 5 Hz (S samples/sec.), and the spectrum }G(~) repeats rene {10s cad/ece), The successive spectra overlap, as shown in Fig. 56.1-44, and the spectrum Gl) is not recoverable from G(w). that is. 9(?) cannot be reconstructed from its samples 9(t) in Fig. $6.1-4c. If the SRinfed Signal is pased through an seal lonpae Biter of bandwidth 5 Hs, che output spectrum i rcs?61-5 0.1-6 61-7 61-8 and the output signal is 10sine (20rt). which i not the desired signal sine °(Srt), In the second case. we tae the Nyquist sampling rate of 10 Hz (Fig, $6.1-4e). The spectrum Ble) consists of back-to-back, nonoverlapping repetitions of #G() repeating every 10 Hz. Hence, G(w) can, be recovered from G(w) using an ideal lowpass Foes cintdit 3 He (Fig, 96.-4f). The output is 10sine (xt). Finally, in the last case of oversampling {sampling rate 20 Hz). the spectrum G(w) consists of nonoverlapping repetitions ‘of #G() (repeating every seen wish empty band between successive cycles (Fig. $6.1-4h). Hence. G(w) tan be recovered from Bw) ith empty Dengiter or even » practical lowpass filter (chown dotted in Fig $6.1-4h). The output it ‘This echeme is analyzed fully in Problem 8.4-1, where we found the bandwidths of y(t), jolt). andu(t) te bs 30 ile SAH: and 5 ie, respectively. Hence the Byeust ates for the three signals are 20 kHz, 10 kHz, and 30 a, respectively (a) When the input to this iter is A(t) the output of the summer is a(¢)— a(¢—7)- This acts as the Input (o the integrator. And, (7). the output of the integrator is: ney [0 8 = er =a ae) = (3) ‘The impulse response h(t) is shown in Fig. $6.1-6a. (b) The transfer function of this circuit is Hl) = rine (“For (8) “The amplitude response of the filter is shown in Fig. $6.1-6b, Observe that the filter is » lowpass filter of bandwidth 22/7 rad/s or 1/T Hz, ‘ppreiwicth 2 Caponse of the circuit isa rectangular pulse, When a sampled signal it appiics a) the input. each Thee generates a recrangular pulse at the output. proportional (9 the Corresponding ‘Sample value, Hence the farnbit Ba staircase approximation of the input as shown in Fig. $6.1-6c. and i) =T ear zw Figure 6.1.6 (a) Figure $6.1-7a shows the signal rcongtrution, fom ite samples using the ratorder ald Grou Eh (2) Figure $6.1 Te shan OF SR aT and contered atthe sampling instant. Te height of he tangle = caval Sample genet a eg signal Consists of straight line segments joining the sample ‘ops (h) The transfer function of this eircut 8 , ; Hw) = Fun} = FLO (&)} a7 (3) Because H(o) is postive forall, i alo represents te amplitude response, Fig. $6.1-7b shows the ‘pues Because (2) ret Se corresponding amplitude response H(4) andthe ideal amplitude response (ownass) Poned (orignal Fzconstruction is shown in Fig. S6.1-7e required fr signal Sonar Geayn required tofmeke 1(é) canal (realizable), Such « delay would caus She (2) A imirie aignal in Fig. 96.1-78 to be delayed by T secs, resoumructed ena oe frat fer in A) then ax shown in Prob. 61-4. it ouuput ie a rectangslt Rule (a) When, the (PF Shown in Fig S8.1-4a. This pulse p() i applied tothe inp of the second iene we a dommes ofthe second Aten lt) — ptt — 1) = u(t) —2udt— 7) w(t 27), which i applied The output of The a niput h(t ofthe integrator is the area under p(t) ~ p(t ~ 7). which. as u(r) = 2n(r = T) tar = IT) dem tut) ~ At — Thule = 7) + (0 BP alt = 27) = me shown in Fig, $6.1-7b. Assume a signal a(t) that is simultaneously timelimited and bandiimited. Let g(a) = 0 for jl > 2x2. ‘Therefore aenneei (gage) = a(e) for B' > B. Therefore from the time-convolution property (3-48) a(t) = ot) = {2B'sine(2"B"t)} = 2B'9(t) + sine(2"B"t) 326.21 6.22 62-3 62-4 6.2-6 4 ney “7 ¢ + t= Cbs Figure $6.1-7 Because g(t) is timelimited. g(t) = 0 for itl > 7. But a(t) is equal to convolution of a(¢) with sinc(27 BY) wich eect (ced, It impossible to obtain a time-limited signal from the convolution of « time-limited signs) with a non-timelimited signal. (a) Since 128 = 27. we need 7 bite/character {(b)For 100,000 characters/second . we need 700 Kbits/second. (a) 8 bits character and 800 kbits/second, (a) The bandwidth is 15 kHz. The Nyquist rate is 30 kHz (b) 65536 = 2°°. so that 16 binary digits are needed to encode each sample. (c) 30000 x 16 = 480000 bits/s. (4) 44100 x 16 = 705600 bite/s. (a) The Nyquist rate is 2x 4.5 x 10° = 9 MHz. The actual sampling rate = 12 x 9 = 10.8 MHz (b) 1024 = 2", so that 10 bits or binary pulses are needed co encode each sample (c) 10.8 x 10% x 10 = 108 x 10° or 108 Mbits/s. If-mp is the peak sample amplitude. then wantization error ¢ 22)Qt2) = Te es = 00 ~ 500 Because the snaximum quantization error is At = 254 = 72. it follows that Merete me L= 500 Because L should be @ power of 2. we choose L = 512 = 2°. This requires a 9-bit binary code per sample The pecs eat x 1000 = 2000 liz 20% above this rte is 2000 x 1.2 = 2400 Ha, Thus, each sina! bas 240) Neat uuona and each sample i encoded by 9 bits, Therefore, each signal uses 9x 2400 = 21.6 kits/acond Fae eignais ave multiplexed. hence. we need a total of 5 x 21.6 = 108 KBits/aecond data bie, Freres and synchronization requires additional 0. '5% bits, that is, 108,000 x 0.005 = 540 bits, yielding a total of 108540 bits/second. The minimum transinission bandwidth is QE = 94.27 kia ‘Nyquist rate for each signal is 200 ‘Hz. The sampling rate f, = 2 x Nyquist rate = 400 Ha Total number of samples for 10 signals = 400 x 10 = 4000 samples/second. Quantization error < 22m = TB Moreover, quantization error = A¢ = 2n¢ = "2 = RB =e b= 400 Because L is a power of 2. we select L = 512 ‘2°. that is, 9 bits/sample. ‘Therefore, the minimum bit rate = 9 x 4000 = 36 kbits/second. The minimum cable bandwidth is 36/2=18 kHz For a sinusoid, SSH = 0.5. The SNR = 47 dB =80119. From Eq, (6.16) ame $0 = 970) was0s) = son =o b= 1028 ecouse L is a power of 2. we select L = 256 = 2%. The SNR for this value of L is Be i220 = g(290)%(08) = 98004 = 49.43487 6.2-8 6.2-10 64-1 For this periodic s(t). each quarter cycle takes on the seme set of amplitude values, Hence, ‘each quarter cycle Foran era weal cvergy, Consequently, we ean compute the power for thie signal by averaging coe seeirriuarter cycle. The equation ofthe fest quarter cycle ax shown in Fig, $6.27 is (0) & 4A/To. where Ais he peak armpliude and Ty is the period of m(t). The power or the mean squared value (energy averaged over a quarter cycle) is mye PAY wo anf My a= $ Hence, SG = At = 4 tena, TF ve schtion i identical to that of Prob. 6.26. From Eq. (616). SNR of 47 dB is a ratio of $0119. So ami(t) eae 3L7(1/3) = 50119 ==> L = 223.87 Because L is a power of 2. we select L = 256 = 2°. The SNR for this value of L is So 2h = 3L2AD = 9(256)7(1 /3) = 65596 = 48.16 dB my Here jv = 100 and the SNR = 48 dB= 31,622.77, From Ea, (6.18) So _ 30? 2 = Garis SNORT 473.83 Becouse L is a power of 2. we select L = 512 = 2°. The SNR for this value of L is ; $2. SOE, oan = as 0r a8 (a) Nyquist rate = 2 108 Ha. The actual sampling rate is 1.5 x (2 x 108) = 3 * 10° Hz, Moreover. L = 256 and 1 = 255. From Ea. (6.18) So ___3L? 3256)” ggg4 = SE > Tage TF ~ zee 7 OM = 5808 SB (b) If we reduce the sampling rate and increase the value of L so that the same number of bits/second is traintained, we can improve the SNR (because of increased L) with the same bandwidth. In part (a). the aaarresreecce is 3x 10" Hz and ench sample is encoded by 6 bits (L = 286). Hence, the transmission rate is 8B x 3x 10° = 24 Mbits/second. If we reduce the sampling rate to 2.4 x 10° (20% above the Nyquist, rate), then for the same transmission rate (24 Mbits/s). we can have (24 x 10°)/(2.4 x 10%) = 10 bits/sample, This results in L = 22° = 1024. Hence. the new SNR is So 3 3(1024)? No ~ Tine + 1) (in256)* tone Clearly. the SNR is increased by more than 10 4B. ‘Equation (6.23) shows that increasing by one bit increases the SNR by 6 4B. Hence. an increase in the SNR. ‘by 12 dB (from 30 to 42) can be accomplished by increasing n from 10 to 12. that is increasing by 20%. (a) From Eq, (6.33) (a)(64.000) 7 = 00785 Ama = 7% sothar 1 342B _ (0.0785)*(3500) () Nom 3p, =~ ByEH000) 1.2 107* (c) Here So = 4 = 0.5. and So os 10" Wg = Taba Tot = 46 10 (4) For uniform distrib mye om So 2.94 x 107 {e) For on-off signaling with a bit rate 64 kHz, we need a bandwidth of 128 kHz. For » bipolar case, we need a bandwidth of 64 kHzChapter 7 a 7.2-1 For full width rect pulse p(¢) = woo) P(e) = J sine (#) . For polar signaling [see Eq. (7.12)] sya) = LOM or sinc? (9% jonteh nae) For on-off case (see Eq. (7.18b)} 2 sy(o)- Ph ate =z 4fo-22)] meme oF snd (B) [1028 £ al 0- 2 sine? (2 We EO? But sinc? ()-0 for ont for alln #0, and =1 for n=0. Hence, hs 5,(0)= Bsine? (B)-Fa0) For bipolar case [Eq. (7.20b)] ‘The PSDs ofthe three cases are shown in Fig. $7.2-1. From these spectra, we find the bandwidths fo all three cases to be Ry Hz. ‘The bandwidths for the three cases, when half-width pulses are used, are as follows: Polar and on-off: 2Ry Hz; bipolar: Ry Hz. Clearly, for polar and on-off cases the bandwith is halved when full-width pulses are used. Howewet for She biolar ease, the bandwidth remains unchanged. The pulse shape has only @ minor influence in the bipolar case because the term sin? (2) ‘Sy() determines its bandwidth. so)122 723 124 pied Tole Sytt2) tt aT oy OF From Fig. S7.2-2, itis clear that the bandwidth is = rad/s of 2R Hz. rs For differential code (Fig. 7.17) ° x zy Ler? u ‘To compute Ry, we observe that there are four possible 2-bit sequences 11,00, 01, and 10, which are equally likely. ‘The product aya. forthe first two combinations is | and is~1 forthe last two combinations. Hence, I ‘Similarly, we can show that Ry = m>1 Hence, S,(o) = (net -(2) sinc? (2) . 4 i Ltn N er [FoF (a) Fig. $7.2-4 shows the duobinary pulse train y{¢) for the sequence 1110001101001010. (b) To compute Ro, we observe that onthe average, half the pulses have ay = Oand the remaining half have a, = Jor -1. Hence, an UPN ay? Moyet tim +) 2 (a1? +SO)J=> Foe ie wld y Xoo} 2 To determine Ry, we need to compute a,ay1- There are four possible equally likely sequences of two bits: 11, 10, 01, 00. Since bit 0 is encoded by no pulse (a, = 0), the product of 444,41 ‘= O for the last three of these sequences. This means on the average 2 combinations have 242441 = Oand only 4 combinations 7134 have nonzero a,a,41. Because ofthe duobinary rule the bit sequence 11 can only be encoded by two consecutive pulses of the same polarity (both positive or both negative). ‘This means ay and.a,,, are 1 and 1 or ~1 and 1 respectively. In either case ax@k+ = 1. Thus, these a combinations have aja,4; 1. Therefore, A in, 1 Aaeo]= 4 4 ‘To compute Ry ina similar way, we need to observe the product ay ay42.. For this we need to observe all possible combinations of three bits in sequence. There ae eight equally Likely ‘combinations: 111, 101, 110, 100, 011, 010, 001, and 000. "The last six combinations have either the first and/or the last bit 0. Hence, a42,49 = 0 forall these six combinations. The first wo combinations are the only ones which yield nonzero o42,,2. Using the duobinary rls, the first combination is encoded by three pulses of the same polarity (all positive or negative). Thus a, and ay,2 are 1 and 1 or ~1 and respectively, yielding ‘a,2p.g = 1. Similarly, because ofthe duobinary rule the frst and the third pulses in the second bit cee tnation 101 are of opposite polarity yielding ayayy2.=I- Thus on the average, 442442 =! for Zrermst for terms, and 0 for 2 corms, Hence, a riot Tole Ina similar way we can show that R, =O > 1, and from Eq, (7.10), we obtain 2 2 5,(0)= Fed +cosah)= Pel on( 8) 2 R= re ee For half-width pulse P(t) = reet(2t/ 7). lhe S,(0) = B sine? From Fig, $7.2-4 we observe thatthe bandwidth is approximately Ry /2 Hz From Eq. (7.32) 38132 13-3 134 13-5 13-6 137 m, Quantization enor 2Y = “Ps 001m, = 12 100 @ Because Lis a power of 2, we select L = 128=27 (&) This requires 7 bit code per sample. Nyquist rate = 2 2000 = 4kHz for each signal. The sampling rate f, = 125% 4000 = 5 KHZ. Eight signals require 8 x 5000 = 40,000 samples/sec. Bit rate = 40,0007 = 280kbits/s. From Eq. (7.32) = (TDR 21228010? = gg yar, 2 2 (a) By =2Ry = Ry = 1S kits. (b) By = Ry => Ry = 3 kbit/s. © Br =a, Hence, 000 = 125 ry = Ry = 4B kbits/s. (@)_ By = Ry => Ry = 3kbitss. (©) By = Ry = Ry = 3 kbitsss, (a) Comparison of P(a) with that in Fig. 7.12 shows that this P(c) does satisfy the Nyquist criterion with @y = 2% 10% andr =1. The excess bandwidth o , = nx108, (b) From Table 3.1, we find pit) =sine? (10%) From part (a), we have ay = 2108 and Ry = 10°. Hence, 7, = 10. Observe that nT) =1 n=O 20 n#0 Hence P(t) satisfies Eq. (7.36). (©) the pulse transmission rate isn Ry =10° bites, a Inthis case At = 1M, Hence, we can transmit data ata rate Ry = 2 MHz. Also, By = 12MHz. Hence, from Eq. (7.32) 6 14P 108) 12x10 =Fex')ar-02 fy = T00kHz. nso, Ae = S00kHz and f, = 700-500 = 200 kHz. fs RB -“_. = 4b_ f, = 500-2002 Hence, = gts = OA and fi =~ fe = 500 200 = 300 kHz. To obtain the inverse transform of P(a), we derive the dual of Eq. (3.35) as follows: s(t—T) 2 G(a)e™ and g{t-+7) 2 G(a)e™ dt+7)+9(t-7) 2 2G(0)cosTo o Hence, ‘Now, P(«) in Eq. (7.34a) can be expressed as noegedaig amt aag)4) G 9‘Using Pair 17 (Table 3.1) and Eq. (1) above, we obtain P(t) Resin (28 yt) + Absine [pea (roags) eo [ea('-35) - [sine (2x Rt) + hse (2m Ry) + sine (25 Rst~") =k ‘sin(2# Ryt) 11 sin(Qa Ryt+ x) 1 sin(2x Rit —#) QR 2 2aRyee 2 2RRl-* . (see 1 sin(2Ryt) Lastet) eRe 2 2K te 2 eRe = Rysin(2 Ret | anh ro _, 2p cose Rytsin Ryt _ Ry cos Ry «, ae Ryi(-4R4?"?) 1-4R 20 sine (x Rt) mnie : ine(s8) [eo werteat os sieda ilo HU) = sine (Rt) sie [-a('-a)] = Sine Rot, sin(a Rot 2) RRR _sine yt _ sine yt. _ sinxRyl ARs ERI -e RI Rit) Hence, 73-9 The Nyquist interval is 7, “yh ‘The Nyquist samples are p (£17) for n= 0,1,2, -..-- From Eq, (7.16), it follows that 0) = A(T) =1 and p(tnTp) = 0 for all other n, Hence, from Eq. (6.10) with 7, = Tp, mip--t, 2° 2h Hi) =sine Ry +e [-a(--z)] (sin a Ryt_ sin wx Ryt | _sin Ryt aR, wR ARe(I- Ry)‘The Fourier transform of Eq. (1) above yields 1 @ 1 @ rondo ste) aman = Lm 2) fes0h 4 , or2 000, These sample values represent the following 4 digit sequence: 1100, or 1000, or 10100, or 1610. Hence the 4 possible correct digit sequences are 1101001001%,x;5%411100 where 1,x,%5% is any of the four possible sequences 1100, 1000, 0100, or 1010. 74-1 S=101010100000111 From example 7.2 Ta(1ep'@D'eD°@r%eD" ep" ep” ep" eo"e..\s R=(1@D°@D*)T T= 101110001101001 R= 101010100000111 = S ‘14-2 $= 101010100000111 [2_]) T=(ieD? eptepepteD ep" ep'e.. Ra(1@D?)T (see Fig. 74-2) T T = 100010000000110 R= 10101G100000111= S Fig. S7.4-2 74-3 S=101010100000111 Tx(1@D@ 5? @D*@D’@D*eD?@D"! ep")s R=(1@De@ D*)T (see Fig. S7-4-3) 7 =110111101001011 R= 101010100000111 = S 6116-1 16-2 (+) ++ .~— + ® Fig. S743, From Eq, (7.45), we obtain e 1 03 -0.077 fo] [-0328 |=] 02 1 03] [t]=| 107 e, } [0002 o1 1 [o} [-o1s 4 @ Bes on () Forpolar case P; = 0(5)= 287107" (i) For on-offcase P, = (5/2) = 0.00621 (Wii) For bipolar case P, = 15Q(5/2) = 0.009315 In the following discussion, we assume Ap= A, the pulse amplitude. (b) Energy ofeach pulse is E, = 4% /2 and there are Ry pulses/second for polar case and & pulses/second for on-off and bipolar case. Hence, the received powers are 2g, # _(oomsy 2 Protat * =1125x10% 2 2 Papo Ah Bh A 0 05625 110% 2 2 2 AT Re 2 #2 05625x10% 2 "2° 4 (©) For on-off case: We require P(e) = 287% 10"? = (4, /20,)- Hence, Ap! 20 = 5 and A, = 100, = 0.003 2 2 Payot -£ 008) =225%10 For bipolar case: 4, P(e) =287x10~ = 150( Ay No) = 5075 Hence A= Ay =5075% 26, = 0.003045 and 2 Papotu = 4- = 231010 For on-off case: ‘ P,=10% sd = fears 2o,) ton 6216-3 14-2 103 on = 107? => A, 2 (4.752107) = 95% 103 For on-off case, half the pulses are zero, and for half-width rectangular pulses, the transmitted power is: 225610 watts. “There is an attenuation of 30 dB, or equivalently, a ratio of 1000 during transmission. Therefore Sp = 10005, = 2256 107? watts For polar case: 4p) 4 10° 2 of 42) > 22 0 475 Ay = 4.752107 on on ? For polar case with half-width rectangular pulse: = Flos 10-3)? = 1128 10-4 wants Sp = (1000}{11.28% 106) = 11.28 107° warts For bipolar case: A, P2106 =159 > AP 24835 and A, = 4835x2109 = 9.67 x10 20,) 20, e For bipolar (or duobinary), half the pulses are zero and the receive power S, for half-width rectangular pulses is 2 4 2 5, = 22 =1(967% 107) = 233810 wans 44 Sp = (1000)5; = 2338% 107? watts Sampling rate = 2 x 4000 x 125 = 10,000 Hz. m, Quantization error = =F = 0.001my => L = 1000 Because L is a power of 2, we select L = 1024=2'°. Hence, n= 10 bits/sample. (@) Each 4-ary pulse conveys log, 4 = 2 bits of information. Hence, we need = 4-ary pulses/sample, and a total of Sx 10,000 = 50,000 4-ary pulses/second. Therefore, the minimum transmission bandwidth is aoe = 25kH2, Ault), 500001125) = 7 = 2 = 3125kHz. © Br (0) Each 8-ary pulse carries log, 8 = 3 its of information. Hence, the bandwidth is reduced by a factor of 3. (b) The amplitudes of the & pulses used in this 8-ary scheme are+A/2, £34/2, +54/2, and £742. Consider binary case using pulses: /2. Let the energy of each of these pulses (of amplitude + 4/2) be Ey. The power of this binary case is Ainary = Eos Because the pulse energy is proportional to the square of the amplitude, the energy of « pulse ad is 47 Ey. Hence, the average energy of the 8 pulses in the 8-ary case above is 63TI 1-4 115 Fy) 221) +23) of z 21E, Ew = ; =21Ey Hence, Peay = Eay pulse rte = 218, x Bb TER Therefore, Peary = TPoinary (0) M-=16. Each 16-ary pulse conveys the information oflog, 16= 4bits. Hence, we need ‘2000 = 3000 16-ary pulses/second. Minimum transmission bandwidth = —— = 1500 Hz. 2 (b) From Eq. (7.32), we have Ry = ier Hence, 3000=-2 By = By = 18008. (a) For pola signaling, Rybits/second requires a bandwidth of Rj Hz. The half-width rectangular pulse of amplitude 4 has energy 2 ‘The power Pis given by P= EsRy ~4n- (b) The energy ofa pulses is A2E,. Hence the average energy ofthe M-ary pulse is Ey = [aes +223) +2185) one M—)P 5] Ma Each Mary pulse conveys the information oflogy M bits. Hence we require only AM ary pulses/second, The power Py is given by Ey (M21 (M-1)? te Py = as Mog, M 310g, M Fo" Talog M - 2Alog2 M Each sample requires 8 bits (256= 2") . Hence: 24,0008 = 192,000 bits/sec. Br = 30kHz reer = (00.000) = = 50,000 M-ary pulses/sec. We have available up to so,000 Mary ypulses/second to transmit 192,000 bits/sec. Hence, each pulse must transmit at least 384 bits.181 7.82 78-3 79-1 => choose 4 bits/pulse => M = 16is the smallest acceptable value (a) Baseband polar signal at a rate of IMbits/sec PSD of Ps and using full width pulses has BW = IMHz. PSK doubles the BW to 2MHz. (b) FSK can be viewed as a sum of 2 ASK signals. Each ASK signal BW = 2MHz. The first ASK signal occupies a band fay +1 MHz, and the second ASK tee fy signal occupies a band f_,+1MHz. Hence, the bandwidth is 2 MHz + 100 kHz =2.1 MHz. tang Ketootaad wwe Fig.S7.6-1 (=) A baseband polar signal at a rate 1 Mbits/sec using Nyquist criterion pulses at r = 02 has a pw =D g, = 2x08 60x10" He PSK doubles BW’ to 1.2 MHz. (b) Similar to Prob. 7.8-1. BWpsx, = 06 MHz + 0.6 MHz+ 100 kHz BWssx = 13 MHz PSD of Fst logy M=2 for M=4. ‘We need to transmit only 05x 10° 4-ary pulses/sec (@) BW is reduced by a factor of 2. BWysx = 1 MHz 3 (b) In FSK, there are four center (carrier) frequencies 4 4, 4 “evs Seas So 404 fa 80 separated by 100 KHz cy ea toy Since ASK signal occupies band f. +05 MHz, the total soo ke bandwidth is 1MHe (05 MH2+0.5 MHz+100 kHz +100 kHz + 100 kHz = 13 MHz. Fig. S7.8-3 mutt) fiz 2800 rotations fre mutt) | Mget) matt)tee ra00 colations Lee Fig. (b) Fig. $7.9-1 Either figure (a) or (b) yields the same result. ‘m(t) has 8400 samples/sec. ‘ma(t), ms(¢), ma(#) each has 2800 ‘samples/sec. Hence, there are a total of 16,800 samples/sec. | : 7.9-2 First, we combine ma(t), m3(t), and mg(t) with commutator speed of 700 rotations/sec. This ‘combined signal is now muplexe with) with a commutator sped of 2800 retationssc,yeling ie OUP of 5600 samples/sec. wot) Comnudalin, speed t K fc, < 2200 rotations \ sec19-3 m4 64-3 kb/s 2 BS oauoc mgt’ 2)-6 kb/s f ' m3H> arebbls 3 er ape . { mie) ae Rb/s (b) S ~ ee 67Chapter 8 Exercises 8.1-1 If aplesiochronous network operates from Cesium beam clock which is accurate to + 3 parts in 1012, how long will it take for a DS3 signal transmitted between two networks to become out of sync if a 1/4 bit length time error results in desynchronization? u Solution: A DS3 bit is transmitted in 1/(44.736-10°) = 2.235336-10" sec. In the worst case, both network clocks will be out of synchronization by 6 parts in 1012. 2.235336-10°/(6:10") = 3922.27 sec/bit or 980.57 sec! % bit 6.1-2 For the bit stream 011100101001111011001 draw an AMI waveform. Solution: sty ge hy Note that typically, for illustrative purposes, the waveform is given as Sia 8.1-3 For the following waveforms, determine if each is a valid AMI format for a DS1 signal. If not, explain why not. a Solution: No. 16 0's violation rl dq 1 re lL po LH r Solution: No. bi-polar violation © od Solution: No. 1's density violation Solution: Yese.1-4.a) You have received the following sequence of ESF framing pattem sequence bits «00110010110010110... Is this a legitimate framing bit sequence in order to maintain synchronization between the T1 transmitter and receiver? Yes___No___ If yes, why? If no, why not? Solution; No. The bit sequence 0011 cannot be in an ESF framing bit sequence. b) The following T1 AMI signal is received: Is this an acceptable T1 signal? Yes___No. a. If yes, explain. b. If no, explain why not (what, if any, DSI standards are violated) and draw on the figure the AMI waveform which would be transmitted by the DSU? Solution: No. 16 0's violation. The 16 0's will be replaced by a pattern of 1's by the DSU. 8.1-5 The signal 110100000000000000001 is received by the DSU ina TI data stream which uses a B8ZS format. Draw the output of the DSU for this signal? The first 1 is already drawn. ‘Show the bit stream which is substituted by the DSU. Solution: 8.1-6 ‘T-1 synchronization at two distant locations is controlled by separate crystal controlled ‘secillators which differ in frequency by 125 parts per million. If the terminal equipment doesn't maintain sync in how many complete D4 superframes will the faster oscillator have generated (at most) one more time slot (8-bit) than the slower oscillator ? Circle the correct answer. 0a)5 b) 10 o) 15 4) 20 ¢) None of the above -if “none”, what is the number of D4 superframes before an extra time slot is generated?, Solution: e) The faster oscillator will generate 125-10°1,544-10° = 193 bits per second more than the slower oscillator. This is one frame/sec = 24.125 time slots. Hence, a time slot difference will be generated in 1/24.125 = 0.04164498 frames or 0.0034704 superframes. 8.1.7 Two plesiochronous digital networks, A and B, utilize Cesium beam clocks accurate to 3 parts in 10!?, The networks are operated by independent long distance companies and are synchronized to each other by means of a UTC signal. a. If a company leases aT! line with D4 framing which is terminated at one end in network A and atthe other end in network B, how often must the networks be resync’d to each other to avoid a framing bit error in the customers T1 signal in the worst case? {You may assume a framing bit error occurs when the two networks are out of sync by 2 1/2 of aT1 “bit time”.} Solution: ATI bit time is 1/(1.544-10° ) = 6.476810” sec/bit. In the worst case, the two clocks would be off by 2:3 = 6 pants in 10"? or 6-10"? errored bits per bit transmitted. Hence, 6.47668-10” sec/bit / 6-10" errored bits per bit = 1,07945-10° seconds per errored bit or 5.39723-10° seconds per errored half-bit. b.UTC operates via GPS satellites which are approximately 23,000 miles above the Earth. How long, in terms of T1 bits, will a correction signal take to be transmitted to the network switches? . Solution: The speed of light is approximately 186000 miles/sec. 23000x2 = 46000miles up and down. 46000/186000 = 0.247 sec 0.247x1544000 = 381850 bits 10Chapter 10 SL 10.1-1 101-2 101-3 10.14 13413 1 52 2 141 1 (&) P(black queen) = “55> = 36 a 12_3 (©) P(picture card) = 5 = 75 (a) Pred card) = @ Psd (© rinse BeS (a) $= 4 oceurs a8 (1,1,2),(1,2,1)2,1,1). There are a total of 6x 6x 6= 216 outcomes, Hence, P(S=4) = 7 (@) $= 9 occurs a5 (12,6), (135), (144), (153), (1.62). 21,6), 22,5) 234) 2:43), 252) 60> 6.15), 3,24), 33.3), 3.4.2), 35.0; 41,4), 42,3) 4.32) 44.05 6s (52.2), 3.1), 6.12), 62,1) P(S=9) 216 (©) S=1Soccurs as (3.6.6) (45,6) (46,5). (5.4.6), (55.5). (5.6.4), (6.3.6), (64,5) (65:4), (6.6.3) 10 P(S = 10) ae Note: There isa typo in tis problem. The probebility that the number/ appears should be ki not, - 6 1a Shia k+2k+3k 44k $5k+6k= 21k k= ro P= |. 2, 3, 4, 5, 6) We can draw 2 items out of 5 in 20 ways as follows: 0:02, 0:03, O:P1, Pa 0,0,, 0:0s, 0;P1, O2P2, 0301, 0302, OsP 1, 0sP2, Pi0s, Piz, Pi0s, P1Ps, P01, P02, P20s, PaP1. All these outcomes are equally likely with probability 1/20. (@ This event E) = 0, F,U0, Fy U0 AUD; Pz U0 AU05 A, UA 0 UA02U AUF 01 U F202 UFOs 2.2 Hence, P(E) = 35° 5 (ii) This event £2 = RRZUAA 201 Hence, P(E2)= 35" 75 (il) This event £5 = 0402 U0403U020, U0203U030; U0302 6 Hence, rE)e£-% (iv) This event Ey = Ep UEs and both £,& £3 are disjoint. 4.2 Hence, P(E.) = P(E2)+ P(E3)= 7975 n101-5 101-6 10.1-7 101-8 Let zo, be the event thatthe first chip is oscillator and xq be the event that the first chip is PLL. Also, let xo, and x, represent events thatthe second chip drawn is an oscillator and a PLL, respectively. Then P(1 osc and 1 PLL) = P20, 4) + Pen 20,) = Hoa lene) Hen con) 3,2) +2 3 “3 Using the notation in the solution of Prob. 10.1-5, we find: 3 © Plrosbta)= 3 & Aabn)=2 (a) We can have("2) ways of getting two 1's and eight 0's in 10 digits 19). OL (2)= a7 45 (wo 1's and eight 0's) = 4(057(05)' = 48(05)" = $3 m (b) Plat least four 0's)= 1-[P(exactly one 0)]+[Pexactly two 0's) +[P(exactly three os)] =(12\(05)!0 = 20 = 5 P(one 0) ('?Xo3) "10452 7 0 Pewo 0's) =('9}05)" “im ae 1024 4S 120 849 P (three 0°s) =('3}03)"° . 1024 5 P(at least four 0's) = 1-{ =~ mize (at least four O's) (S+4 “i008 (a) Total ways of drawing 6 balls out of 49 are 49), 49! 8 (2)* aa aa 7 13983.816 Hence, Prob(matching all 6 mumbers)= ar (b) Tomatch exactly 5 number means we pick 5 of the chosen 6 numbers and the last number can be picked from the remaining 43 numbers. We can choose 5 numbers of our 6 in($) = 6 ways and can choose ‘one number out of 43 in(‘?) = 43 ways. Hence, we have 43 x6combinations in which exactly 5 numbers match, Thus, ‘(matching exactly 5 numbers) = BS = 1845x107 © -Tomatch exactly 4 numbers means we pick 4 out of the chosen 6 number in(§) = 15 ways and choose 2 out of the remaining 43 numbers in( 4?) = 903 ways. ‘Thus there are 15x 903 ways of picking exactly 4 numbers out of 6 and a 15x 903 4 = 1Sx203 9.686% P(qmatehing exactly 4 numbers) = 757g = 9.686% 10(@) Similarly, we can pick three numbers exactly in( $4) = 20x 12341 = 246820 ways. Hence, (matching exactly 3 numbers)= a =001765 10.1-9 (a) Let f represent the system failure. Then P(7)=(1-0.01)'° = 090438 P(f)=1- P(7) = 00956 (&) P(f)=099and P(s) = 001 If the probability of failure of a subsystem s; isp, then P(7)=(1- p)” oF 099 (1- p)" = P= 00010045 10.1-10 If f represents the system failure and f,, and f, represent the failure of the upper and the lower paths, respectively, in the system, then: O) PUA) = Paha) = Ueda) = [Pell PUfy)= 1 P(J,) = 1-(1-0.01)'° = 0.0956 and P(f) =(0.0956)* = 0.009143 Reliability is P(7) = 1- P(/) = 09908 © P(7)=0999 P(s) = 1-099 = 0001 P(f,) = 0001 = 0.0316 P(j,) =(I~P)'° = 1-00316= P = 0.003206 10,1-11 Let P be the probability of failure of a subsystem (5, or 52) For the system in Fig. ‘The system fails ifthe upper and lower branches fail simultaneously. The probability of any branch not failing is (1- P\(t- P) =(1- P)?.. Hence, the probability of any branch failing is (I~ Py. Clearly, Py, the probability ofthe system failure is P -[-0-97 ft -(- pyaar? P«l For the system in Fig. b: ‘We may consider this system as a cascade of two subsystems xy and xp, where xs the parallel combination of's and s,and is the parallel combination of sp and sp. Let P(x;) be the probability of failure of ‘Then Py(xy) = Pr (xa) = P? “The system functions if neither x, nor x fails. Hence, the probebility of system not failing is(1- P?)(1- P?), Therefore, the probability of system failing is Py =1-(1- P21 P?)=2P?— Paap? Peel Hence the system in Fig, alas twice the probability of failure of the system in Fig. b. B10.1-12 There are (52) = 2598960 ways of getting 5 cards out of 52 cards. Number of ways of drawing $ cards of the same suit (of 13 cards) is(1?)= 1287. There are 4 suits. Hence there are 4 x 1287 ways of getting a flush. Therefore, 4x 1287 = ARU287 «19808 «1077 P(ftush) = eoeo 7 1980! 10.1-13 Sum of 4 can be obtained as (1,3), (2,2) and (3,1). The two dice outcomes are independent, Let x, be the ‘outcome of the regular die and x, be the outcome of irregular die. Pray (13) = Py (1)Pay (3) = be Pax, (2:2) = Pry (2): (2)= 5 atxoe 1 8 0 1 Pan Ql) = Pa VP ew z i 1 36°12 P)= Pal Fe Ra tale ay (3:2) + Pays (4a!) 1 ald wdyorteteteosteted oe 36" 6 6 2 Therefore P,(4) = Similarly, 101-14 B= ABUA‘B (a) = P(4)P(a}4) + P(4°)>(@]4°) -(Sla(ala)-% Aap) = a 5 a sequence of S digits can occur in($) = lOways. The probability one such 10.1-15 (a) Two I's and three 0” sequence is P= (08)*(02)? =0.00512 ‘Since the event can occur in 10 ways, its probability is 10x 0.00512 = 0.0512 (b) Three 1°s occur with probabitity($)(08)°(0.2)? = 02048, Four 1's occur with probability (3 )(08)*(02)! = 0.4096 Five 1’s occur with probability ($){08)°(02)° = 03277 Hence, the probability of at least three 1°s occuring is P = 0.2048 + 0.4096 + 03277 = 0.9421 10.1-16 Prob(no more than 3 error)= P(no error) + P(1 error) + P(2 error) + P(3 error) = (1 A) (1) re A) + (I)R2I- BY +(P) 2-2)” =(1-1002,) +1002, (1-992) +49507,2(1-98P,)+161700P,°(1-977) 4101-17 Error ean occur in 10 ways. Consider case of error over the first link (correct detection over every link) = (1- A.)(1— }.-(1- Fo) Pg =1-P=1-(1-AXI-A)-l- Ao) = 1-[1-(A + Fat. Ao) + higher ordes terms} At Atty A< e| 3 2 j= Fig. $10.2-4 alf-wave rectifier, y assumes only positive values. So P(y <0) = 0. Since this is Hence, Fy(») = 0 (fory <0) and Ply <0") =. Hence, F,(0") = 102-8 x isa gaussian rv. with mean 4 ando =3 Hence, @ P(x24)=o 44) ~ 019-08 im (2) © rtxz)= of 254) =1-of$) =1-008176~ 09083 I = a © Poez2yed 1-Q(2) = 1- 0.02275 = 09773 ° 4 * Fig. 102-5 10.2-6 (a) From the sketch it is obvious that x is not gaussian. However, it is a unilateral (rectified) version of Gaussian PDF. Hence, we can use the expression of Gaussian r.v. with a multiplier of 2, For a gaussian r.v. LP withe, = Py)= Eee with oy =4 (b) Hence, @) rixz1)=27(y21)=26{2) 0802s oat Ri ai) mrexaa)-anters2)=dd{2)-f)]-ouss Oo x> (© If we take a Gaussian random variable y . eve Fig. $10.2-6 Py) ae and rectify y (ll negative of y multipled by -1), the resulting variable is the desired random variable x 6