PROS a 24 (@) Find the Fourier transform of the half-cosine pulse shown in Figure P2.1a. ‘Apply the time-shifting property to the result obtained in part (a) to evaluate the spectrum of the half-sine pulse shown in Figure P2.10. (©) What is the spectrum of a half-sine pulse having a duration equal to a7? ‘What is the spectrum of the negative half-sine pulse shown in Figure P2.1¢? (©) Find the spectrum of the single sine pulse shown in Figure P2.Md. ) @ Figure P2.1 2.2. Evaluate the Fourier transform of the damped sinusoidal wave a(t) = explt)sin(2afet)u(t) where u(t) is the unit step function. 2.3 Any function g(?) can be split unambiguously into an even part and an odd part as shown by (0) = Belt) + Bol) ‘The even partis defined by seli)= Hie?) + 8-0] and the odd part is defined by lt) = Fle) ~ -0) (a) Evaluate the even ans odd parts of fined by a) Aare G ca 5) o parts of thy of these 00 Pa (&) What are the Fourier wansfors oft pulse? ast lewy Figure P2.4 ne Fourier transform of the frequen termine the inves 2.4 Determine the py the amplitude and phase spect show function GL) defined by in Figure P24. 2.5. The following expression may be viewed as an aPniny ewe representation of a pulse with finite rise time: Vor a 2 where itis assumed that T >> t. Determine the Fourier trans Of a(1). What happens to this transform when we allow r tk come zero? Hint: Express g(t) as the superposition of two sig fone corresponding to integration from ¢ — T to 0, and the a from Oto + T. 2.6 The Fourier transform of a signal g(#) is denoted by @ Prove the following properties of the Fourier transform: (a) If areal signal g(t) is an even function of time 1, the Fee transform G(f) is purely real. If a real signal g(t) isa, function of time 1, the Fourier transform G(f) is ¢ imaginary. f (b) ( ray =(Z) orn where Gf) is the nth derivative of G(/) with sre” ( ad3.7, The Fourier transform GCF) of a signal g() is bounded by ‘folowing three inequalities: ont = [ tecone dg(t) vewccnt = f° |e |e a(t) waanraunis f° [8 Ja where itis assumed that the first and second derivatives of (1) ‘exist. Construct these three bounds for the triangular pulse shown in Figure P2.7 and compare your results with the actual amplitude [esta ofthe pulse. | a) j 7 Figure P2.7 f bs Prove the following properties of the convolution process: @) ‘The commutative property: | u(t) * ga(t) =a2(0) * ail) (0) The associative property: | Bilt) * [gale) * gs(0)) = [eile * 8200] * asl) The distributive property: silt) + [galt) + a3(t)] = eile) * g2(6) +8i(®) * 830) 2 Comer the convolution of two signals 6) and #2 d Flee) * gan] = [Sev] + gl) ) [ta + aod [f. ates] « #2) 2.10 A signal x() of finite energy is applied to a square-law de- vice whose output y(t) is defined by =F) The spectrum of x(t) is limited to the frequency interval —W = J W. Hence, show that the spectrum of y(0 is limited to ~2W = £5 QW. Hint: Express y(0 as x( multiplied by itself. 2.11 Evaluate the Fourier transform of the delta function by considering it as the limiting form of (1) a rectangular pulse of unit area, and (2) a sine pulse of unit area. 2.12. The Fourier transform GC) of a signal g(0 is defined by 1 f>0 an=y7 47° 0 s<0 Determine the signal (0. 2.13. Show that the two different pulses defined in parts (a) and (8) of Figure P2.1 have the same energy spectral density: 4A272c0s2(nTf) (D= ane iy 2.14 (a) The root mean-square (rms) bandwidth of a low-pass signal of finite energy is defined by 5 ® ff fiewtar [ louear where IG(f)F is the energy spectral density of the signal. Correspondingly, the root mean-square (rms) duration of the signal is defined by Wons = © wn Pig at [- wear a Using these definitions, show that 1 TimaWens = 3 Assume that |g(1)| 0 faster than 1,/{¢] as |t| + 00. Considers Gaussian pulse defined by a(t) =exp(-a?) ‘Show that, for this signal, the equality 1 Toe Were = Toms = (b) can be reached.2.15 Let x(r) and y(s) be the input and output signals of a linear time-invariant filter. Using Rayleigh’s energy theorem, show that if the filter is stable and the input signal x(r) has finite energy, then the output signal y(v) also has finite energy. That is, given that x [ moPar< x then show that “* [- pwra 2W, show that m(1) can be obtained from the square-rooter output v3(t). (0 F Low-pass |, [” Square- 5) ——S} Squarer | Lopes |) Sore eae) (= s2() (1) = Vg) Figure P3.7 3.8 Consider a message signal m(¢) with the spectrum shown in Figure P3.8. The message bandwidth W = 1 kHz. This signal is applied to a product modulator, together with a carrier wave ‘Ac cos(2nf-t), producing the DSB-SC modulated signal s(¢). The modulated signal is next applied to a coherent detector. Assuming perfect synchronism between the carrier waves in the modulator and detector, determine the spectrum of the detector output when: (a) the carrier frequency f- = 1.25 kHz and (b) the carrier fre- quency f- = 0.75 kHz. What is the lowest carrier frequency for emWhich each component of the modulated signal s(t) is uniquely de- where A- cos(2f-t) is the carrier wave and mit) jy then termined by m(a)? signal. This modulated signal is applied to a square °\ : characterized by « mp 5 y(t) = 2(1) The output (0) is next applied t0 a narrow-band fier win, band amplitude response of one. midband freqsny 4% bandwidth Af. Assume that Af is small enough to teat :® trum of y() as essentially constant inside the passhand yi “WOW (a) Determine the spectrum of the square-law device cua, ji (b) Show that the filter output v2) is approximately sing Figure P3.8 given by 3.9 Figure P3.9 shows the circuit diagram of a balanced modu- lator. The input applied to the top AM modulator is m(t), whereas u(t) = FEAL cos(dnfer) that applied to the lower AM modulator is mi); these two modu- . lators have the same amplitude sensitivity, Show that the output ‘where & isthe energy of the message signal m( s(t) of the balanced modulator consists of a DSB-SC modulated signal 3.13 Consider the quadrature-carrier multiplex sjsem Figure 3.16. The multiplexed signal s(0) produced atthe tansee mo —>| aM ao output in Figure 3.16a is applied to a communication chan: modulator transfer function H(f). The output of this channel is in tum. appl to the receiver input in Figure 3.166, Prove thatthe condition A, 608 (2H /,t) HL+N=H(f-f), Osssw Oscillator a is necessary for recovery of the message signals m,(t) and mt 2 f the receiver outputs; fis the carrier frequency, and IW is the no Aca (2Rfet) sage bandwidth. Hint: Evaluate the spectra of the (wo recci outputs. =m} Att, FBO 3.14 Suppose that in the receiver of the quadratue-cirs ‘multiplex system of Figure 3.16 the local carrier available fr ‘modulation has a phase error @ with respect to the carier st igure P3.9 Used in the transmitter. Assuming a distortionless communica channel between transmitter and receiver, show that this pb error will cause cross-talk to arise between the two demodls signals at the receiver outputs. By cross-talk we mean that 2 P* tion of one message signal appears at the receiver output belt ing to the other message signal, and vice versa. 3.10 Figure 3.10 shows the circuit details of the ring modu- lator. Assume that the diodes are identical and the transformers are perfectly balanced. Let R denote the terminating resistance at the input end and output end of the modulator (assuming ideal 1:1 transformers). Determine the output voltage of the modulator ii for each of the two conditions described in Figures 3.108 and 3-15 A panicular version of AM stereo uses quaatue a 3.10c. Hence, show that these two output voltages are equal plexing. Specifically, the carrier A, cos(2a/.1) is used to mods! magnitude and opposite in polarity. the sum signal , mi(t) = Vo + mit) + mt) 3.11 A DSB-SC modulated signal is demodulated by appying here Y isa de offset included for the purpose of transit seemed the carrier component, m is the left-hand audio signal, and (a) Evaluate the effect of a frequency error Af in the local car- _ is the right-hand audio signal. The quadrature carrier A, sin(2 rier frequency of the detector, measured with respect to the is used to modulate the difference signal ‘carrier frequency of the incoming DSB-SC signal (b) For the case of a sinusoidal modulating wave, show that be- male) = mile) — me) i cause of this frequency error, the demodulated signal exhi- (a) Show that an envelope detector may be used to rec0¥" bits beats at the error frequency. Illustrate your answer with sum m,(t)+mj(0) from the quadrature-multiplexed si a sketch of this demodulated signal, How would you minimize the signal distortion produ the envelope detector? 2 3.12. Consider the DSB-SC signal () Show that a coherent detector can recover the differ m()— m0). A) = Ae cos(ft)n(e) (©) How are the desired m() and m,(0) finally obtained?= 3.16 The single tone modulating signal mt) = Ay cos sett generate the VSB signal MO) = Am cOs(2Rfat is cos|2: 1 atm 60s[22(f. fn)! + 3AmwAe(I = a)eos|2R( J — fa) (0 where a isa constant, less than unity, representin . Fepresenting the attenuatio ofthe upperside frequency. eae fa) we represent this VSB signal as a quadrature cart inultplex “ s(t) = What ismyn? (by The VSB signal, plus the carrier A, cosaf. is passed through an envelope detector. Determine the distortion pro- dluced by the quadrature component, ms) fo. What is the value of constant a for whieh this distortion 1s worst possible condition? cm 1)e0s(2nfet) + Acoa(e)sin2aft) 3.17 Using the message signal | mn aA | seemi ani Aste the modaed ves forte followin soo of mls {@) Amplitude modulation with 50 pereent modulation (0). Double sideband-suppressed carier modulation. 3.48 The local oscillator used for the demodulation of an SSB Signal st) has a frequency error Af measured with respect 10 He tamer frequency f_ used 10 generate s(1). Otherwise, there is Por” fect gnehyonism between this oscillator in the reeelver and tHe srcillzer spplying the carrer wave in the transmitter, Evaluate the demodulated signal forthe Following two situations: ts of the upper sideband only. ‘the lower sideband only. (a) The SSB signal s(1) con (0) The SSB signal s(2) consists o 3.19 Figure P3.19 shows the block diagram of Weaver's method for generating SSB modulated waves. The MessaBe (modulating) Seal mth Timited to the frequency band fo =I fi. The auxiliary cartier applied to the oduct modulators has a frequency fy, which lies at the band, as shown by the first pair of pt ‘center of this fo entical ‘The low-pass filters in the upper and lower branches are identical, ‘each with a cutoff frequency equal t0 (fi ~ fay2. The caiet - plied to the second pair of product modulators has a frequeney Je tht is greater than (fy — fg/2- Sketch the spect the vai ‘Points in the modulator of Figure P3.19. and hence show tha (0) For the lower sideband, the consibutons of I upper ane lower branches are of opposite polaris and ; dat the cod Tower sideband is SUPPrESSSS lulator output, the lowe! bathe UPPeT ‘and and by adding © ower branches are of the same polarity them, the upper sideband is transmitted: (©) How would you modify the modulator of Figure P3.19, $0 that only the lower sideband is transmitted? no SR ne" | Ss AO am au exe fe He} t t se af ena Figure P3.19 3.20 (a) Consider a message signal m(t) containing frequency com- ponents at 100, 200, and 400 Hz. This signal is applied t0 an SSB modulator together with a carrier at 100 kHz, with only the upper sideband retained. In the coherent detector used to recover m(), the local oscillator supplies a sine wave of fre- {quency 100.02 kHz, Determine the frequency components of the detector output Repeat your analysis, assuming that only the lower sidebana ted. ) 3.21 The spectrum ofa voice signal mi is zero outside the in- anal f= [/1 fh In order to ensure communication privacy. this Signal applied toa scrambler that consists ofthe following e3s- aaamot components: a product modulator, a high-pass filter, seceond prodiet modulator, anda How-pass filter. The earir wave Applied tothe fist prod modulator has a frequency eal 10 fo aireas that applied to the second product modulator has a fre- ueney equal to f+ Je: oth of them Rave unit amplitude, The High pass and low-pass filters ave the same cutff Frequency af Assume that fc > So Derive an expression for the scrambler output s(0), and sketch its spectrum. Show that the original voice signal m(t) may be recovered from s(t) by using an unscrambler that is identical to the unit described above. @ ) 3,22 A method that i used for carir recovery in SSB own. econ systems involves transmitting two pilot frequencies 1S appropriately positioned with respect t0 he transmitted sideband. strated in Figure P3.22a for the case when only the This is il y Tras sideband i tasted In this case the pilot feavences fi ‘and fyare defined by Sake Woh and faker of 99 —— aan3.24 Consider a multiplex system in which four input signals mj(t), mo(t), mg(t), and ma(z), are respectively multiplied by the carrier waves [cos(2zfut) + cos( (2nfst)] [cos(2nfut + a1) + cos(27fot + B)| [cos(2nfyt + 42) + cos(27fit + B3)] [cos(2nfut + fs) + cos(27fot + B)| and the resulting DSB-SC signals are summed and then trans- mitted over a common channel. In the receiver, demodulation is achieved by multiplying the sum of the DSB- SC signals by the four carrier waves separately and then using filtering to remove the unwanted components. (a) Determine the conditions that the phase angles %, %2, 43 and Bi, B>, Bs must satisfy in order that the output of the kth de- modulator is m,(t), where k = 1, 2, 3, 4. (b) Determine the minimum separation of carrier frequencies fo and f, in relation to the bandwidth of the input signals so 4 to ensure a satisfactory operation of the system.