6.14 A Fgura 6.53 mostra uma constlo de nas 240-QAM, a qual pod ser visa como uma forma amplaa decontelaio QAM cna Problem 6.14 Apna saa Cross Conshd. mas ip ay Squee i 5 4 conse. STAT al Tea a ee ee Quarter superconst lam f V.34 modem epv 3B » xe yt 315) Bs DD oe tot pw egos ely ge y Ps peeereals ep gp eye Wo sea B esa Hes Bue pweeDeiles eRe a [| PPE TUT e eer ay Se ee ee er a 2 se eens ge a aS ag ys weg ap fig igh ay mp ae ay 233 211 200 196 204 203 ye eS a x 3 FIGURA 6.53. Superconstlagio em quatro quarts de modem V.34 com 240 pontos de sna A superconstelao com- pleta¢obtida combinando-se as verses destes pontos giradas em 0, 90, 180 c 270 graus,(Extaido de Forney era, 1996, com autorzago do IEEE)6.27 (a) Esbooe as formas de onda das components em fsec em quadratura do sinal MSK em resposta seqiéncia bindra de entrada 1100100010, (b) Esboce a propria forma de onda MSK pata a seqiénclabinria espeifcada na pare (a). wa) AA ee mo ——— es moa tgs dt)6.30 Resuma as similaridades diferencas ente os sinais MSK padtéo e MSK filrados por respostas gaus ‘slanas, Problem 6.30 Comparing the standard MSK and Gaussian-filtered GMS signals, we note the following: (j) Similarities + Fora given input sequence, the waveforms produced by the MSK and GMSK modulators, are very similar, as illustrated by comparing the GMSK signal displayed in the solution to Problem 6.29 and the corresponding MSK signal displayed in Fig. 6.30 of the textbook for the input sequence 1101000, + They both have a constant envelope. (b) Differences ‘The use of GMSK results in a slight degradation in performance compared to the standard MSK for a time-bandwidth product VT, = 0.3. However, the GMSK makes up for this loss in performance by providing a more compact power-spectral characteristic,TA. Dados, a uma taxa de 6 kits, devem ser tansmitdos em uma lin privada com lrpura de banda de 4 Kz, usando pulsos que stisfram o citi de Nyquist, Determine o mimo valor do ator de dca mento rquepodeser usado, ‘ (132 131 FromEg (732) on (eat Alargua de banda de Pf) em het & 0,5 + fem que f lengua de banda além da minima largora de banda R/. Sea ra sui entre o excesso de largura de banda J, ¢a minima Jargura de ban- da (obrica R2: . excesso de largura de banda ee Fala laura de banda tedrca er © O5Ry = Th (731) Observemos que f,ndo pode ser maior que Ry/2, Osrsl (7.32)7.37 Fout-level iransmission atthe pulse rate of 2 Mbit i to be transmitted by means of Nyquist fist criterion pulses with P(f) shown in Fig, P7.3-7. The frequencies f, and fs (in Ha) of this spectrum are adjustable, The channel available for transmission of this data has a bandwidth of 650 kHz. Determine fj and fy, and the roll-off factor Figure P.7.3-7 rt Os 3 A ! / Mite > 1.35 Um pulso p(t) cujo espeeto Pj) 6 mostrado ma Fig, ,7.55 satisfaz o crtério de Nyquist, Se f= 08 Milze;= 1,2 Me, determine a minima taxa em que dads binéio podem se transmits por esse pulso usando ocritrio de Nyquist Qual o valor do fator de deceimento? Figura E75 1 fe08 1 fed (ie od 14 itis i Hence we can tanita a eR «2 MB Algo, By =12MHz. Hen, fom Ea (7.2) satis pei ar=te771 Em um esquema PAM com M= 16, ' (a) Determine a minima largura de banda nevesséria para transmitir dados a uma taxa do 12,000 bites com ISI zero, (b) Determine a largura de banda de transmissto se pulsos que satisfazem o critério de Nyquist com fator de decuimento r= 0,2 forem usados para transmitic os dados, wl (at) M26. ach 16ay pulse conveys the infomation flog 16=4 bis Hence, we need 2 ay pleted. vious sisi bite 21S 2 (b) From Eq, (132) wehavey = Br. Hence, sos p> B=6.9 A computer puts out binary data at the rate of $6 kilobits per second. The computer output is ‘transmitted using a baseband binary PAM system that is designed to have a raised-cosine pulse spectrum, Determine the transmission bandvwidch required for each ofthe following roll-off factors, (a) « = 0.25 (b) = OS (6) a= 075 (a= 10 Problem 6.9 The bandwidth B of.a raised cosine pulse spectrum is 2By -/,, where By = 1/27), and f, = By(| a) Thus B = Bo(1 + @), For a data rate of 56 kilobits per second, By = 28 kHz, (a) @ = 0.25, B = kHz x 1.25= 35 kHz (b) a=0,5, B = 28kHzx 1.5=42 kHz (c) a=0.75, B = 28x 1,75=49 kHz (d) a=10, B = 28x2=56kHz6.12 You are given a channel of bandwidth 3.0 KHz. The requitement is to transmit data over the channel at the rate of 4.5 kilobits/s using binary PAM. (a) What is the maximum roll-off factor in the raised-cosine pulse spectrum that can accom- modate this data transmission? (b) Whatis the corresponding excess bandwidth? Problem 6.12 We are given the following specifications: B, =3 kh i = 45 kilobits/s (a) The transmission bandwidth is related to the roll-off factor by the formula (see Eq. (6.21)) By = Bl +a) w where B= QT) Therefore, with (1/T)) = 4.5 kilobits/s, we have By = 225 kHz Hence, solving Eq, (1) forthe roll-off factor, we get 2 r = OF, (b) The excess bandwidth is defined (see Eq, (6.22)) f, = aB,8.36 The power spectral density of a random process X(t) is shown in Fig, 8.27. (a) Determine and sketch the autocorrelation function Ry(r) of X(t). (b) What is the dc power contained in X(t)? (c) What is the ac power contained in X(t)? (d) What sampling rates will give uncorrelated samples of X(t)? Ate the samples statistically independent? Saf) af) 10 % ¢ TL Fteure 8.27 Problem 8.36. Solution (a) Using the results of Problem 8.35, and the linear properties of the Fourier transform R(r)=1+ Ysine*(f,z) (b) The de power is given by power centered on the origin de power = tin 5, (Aaf = tin a(fiaf =l (c) The ac power is the total power minus the de power ac power = Ry (0)- de power =R,(0)-1 =f (d) The correlation function Ry z) is zero if samples are spaced at multiples of 1/fo.8.37 Consider the two linear filters shown in cascade asin Fig, 8.28, Let X(t be a stationary process with autocorrelation function R,y(r). The random process appearing at the fist filter output is V(¢) and that at the second filter outputs Y(t). (a) Find the autocorrelation function of V(z). (b) Find the autocorrelation function of (2), va) Xi— bi) — hi) — ye) Ficure 8.28 Problem 8.37. Solution Expressing the first filtering operation in the frequeney domain, we have WA)= HAD) where H,(f) is the Fourier transform of f(t), From Eq, (8.87) it follows that the spectrum of Mi is SA =| NSA) By analogy, we have SA AVY) =P PS.) Consequenily, apply the convolution properties of the Fourier transform, we have R(=sO* ORD where * denotes convolution; g>(i) and gi(i) are the inverse Fourier transforms of |H:(f)° and |Fh(f), respectively.64. A.componente de sinal de um sistema PSK coecente€definida por Ai) = Aksen(2mft)t AV = cos2nft) onde 0 $$ T, o sinal mas corresponde ao simbolo 1 eo snal menos, ao simbolo 0. O pritnciro fermo representa uma componente de portadora inclufda com a finalidade de sincronizar 6 receptor com 0 transmissor, (a) Desenhe um diagrama de espaco de sinal para o esquema aqui descrito; quais observagdes voce pode fazer a respeito deste diagrama? (b) Mostre que, em presenga de rufdo branco gaussiano aditivo com valor médio igual a zero ¢ den- sidade spectral de poténcia Ny/2,a probabilidade médias de ero é onde (0) Suponhamos que 10% da poréncia de sinalrransmitido sejam alocados & componente da porta- dora, Determine a E/N, necesstria para realizar uma probabilidade de erro igual a 10. (d) Compare esse valor de E/N, com 0 que é necessério para um sistema PSK convencional com a mesma probabilidade de erro,65 (a) Dadua seqlénciabiniria de entrada 1100100010, esboce as formas deonda das componente em fase cem quadedura de uma onda modula obtidausandose o QPSK based no conjuno de sal da Figura 6.6, (b) Fsbceaprépiaforma de onda QPSK para sen bind end spcicda no ite), 6.14 A Fgura 6.33 moa uma cone de sins 240-QAM,a qual pod se visa como uma forma ampliada de constelago QAM cra, (a) Identifique a parte da Figura 6.53 que é uma constelacéo QAM quadrada. (b) Desenvolva o iuem (a) para idencficar a parte da Figura 6.53 que é uma constelagio QAM cru- rad (©) Entao, identifique a parte da Figura 6.53 que é uma extensio da constelagio QAM cruzada. 6.15 Deine edu de pura de banda deans a nerds mead 256 QAM em comparacio com a H-QAM, 6,20 Os vtoresdesinal se, so usados para repesentar os imbolos bins |e, respctivamente, er 'um sitera PSK biti coerent, ( ecepror decide em favor do simbolo | quando rot X59X5, one's 0 produto intero do vero obsevago x do vetr sina §,,onde '= 1, 2, Mosre que ex saregra de decisfo€equialente condo > x, onde, soos elementos do vtorobservagio. X, Suponha que os vecores snl se, tenham energa igual,6.21 Um sister ESK transite dados bindros& taxa de 2,5 x 10" bits por segundo. Durante o curso da transmssio ruido branco gawsiano com valor médio igual a zero ¢densidade espectal de poténcia 10" WH éadicionado ao sia, Naauséncia de rafdo, a amplitude da onda senda eebida para 0 digito | ou 0 1 mY, Determine a probabilidade media de ero de simbols paa as soguintescon- figuragées de sistema: (a) FSK bindtio coerente (b) MSK cocrenre (c)_FSK binatio no-coerente 624 Crie um diagrama de locos para a geragio do sinal FSK de Sunde 4(d) com fase continua usando a representa dada a Equi (6.104), a qual €reproduid aqui 2H, [at 25, (mt = bod — leos(dnfi)? 2 4 (oft "a fh "i sen(2aft)1.24 Um ruido branco 1) com densidade espectal de poténcia Nj/2 éaplicado a um filtro passa-baixas Buttenvorth de ordem n, caja resposta ¢ definida por 1 l4f]2= ——, bev] (a) Determine a largura de banda de ruido equivalent para ess filtro passa-baxas, (Leia 0 Apéndice 2 para ober a defini de largura de banda de ruido equivalent:) (b) Qual é valor limitador da largura de bands de ruido equivalenre quando 1 se aproxima do inf nito? ble 1.24 (a) The noise equivalent bandwidth is V2 - 2 sf mE at 1H(O)1? Le if of Fe + (4/4) zt af 01+ (f/f) ae, * Bn sinta/en) fy * Sine (17eny (b) When the filter order n approaches infinity, we have 1 W din StmecT7eny1.23. Considere um processo de rude branco gaussiano de valor médio igual a zero c densidade espectral de poréncia Nj/2 que seja aplicado & entrada do filtro RL passa-altas mostrado na Figura P1.23. (a) Encontre a fungio de autocorrelacio ¢ a densidade espectral de poténcia do processo aleatério na saida do filtro. (b) Quis séo a média e a variéncia dessa saida? k Bnerada £2 Saida ‘Problem 1.23 G) The poser spectra density of the notse at the filter output 4s given by ry = Oy seen Si = 2 Na ger U % 2 lo (nny (2 = a Veet L/R seu- (b) The mean of the filter output is equal to H(0) times the mean of the filter input. The process at the filter input has zero mean. The value H(0) of the filter’s transfer function H(f is zero. It follows therefore that the filter output also has a zero mean. ‘The mean-square value of the filter output is equal to Ry(0). With zero mean, it follows therefore that the variance of the filter output is oy = Ry(0) Since Ry(t) contains a delta function &(¢) centered on t = 0, we find that, in theory, oy” is infinitely large.igual a zero, fungio de autocor- s que X(0) seja um processo estacionétio com valor mi 40 R,{r) e densidade espectral de poténcia S,('). Somos solicitados a encontrar um filtro linear esposta ao impuls H(), tal queasaida de filtro renha as mesmas caracteristicasestatstcas de X(0) entrada é um ruido branco com densidade espectral de poréncia N,/2. (a) Determine a condigéo a que a resposta ao impulso M(#) deve sarisfazer para cumprir essa exigéncia, (b) Qual éa condicéo correspondente sobre a esposta em freqiéncia Hf do filtro? (2) The autocorrelation function of the filter output is Ry(t) = LL bey) berg) Ryltoryeta) dry dt Sinee Rylt) = (Wo/2) 6(x), we find that the impulse response h(t) of the filter must satisfy the condition: Ryle) 2 L Lary) Brg) 6loonyttp) ary erg Ye Fy Mae) mrp) dtp (b) For the filter output to have @ power spectral density equal to Sy(f), we have to choose the transfer function H(f) of the filter such that Sy(f) or (RCE) =Considere um pulso retangulardefinido por ()= A, OS#sST RETO, ‘caso contrério O que se prope & aproximar o filtro casado para gt) por um filo passa-baixas ideal com largura de banda Bi a maximizacio da relagio sinal-ruido do pulso méximo é o principal objetivo. (a) Determine o valor étimo de B para o qual o filtro pasa-baixas idal fomnece a melhor aproxima- 60 para o filtro casado, (b) Em quancos decibes filtro passa-baxas ideal é pior do que o filtro asada? Problem 4.3 Ideal low-pass filter with variable bandwidth. The transfer function of the matched filter for a rectangular pulse of duration t and amplitude A is given by Hog(O = sine(fDexp(-jntT) @ The amplitude response |H,p(f) of the matched filter is plotted in Fig. 1a). We wish to approximate this amplitude response with an ideal low-pass filter of bandwidth B. The amplitude response of this approximating filter is shown in Fig. 1(b). The requirement is to determine the particular value of bandwidth B that will provide the best approximation to the matched filter. We recall that the maximum value of the output signal, produced by an ideal low-pass filter in response to the rectangular pulse occurs at t = T/2 for BT < 1. This maximum value, expressed in terms of the sine integral, is equal to (2A/x)Si(xBT), The average noise power at the output of the ideal low-pass filter is equal to BNg. The maximum output signal-to-noise ratio of the ideal low-pass filter is therefore (2A/n)Si2(QBT) @) (SNR), No ‘Thus, using Eqs. (1) and (2), and assuming that AT = 1, we get(SNR), 3 (SNR), x2BT (xBT) This ratio is plotted in Fig. 2 as a function of the time-bandwidth product BT. The peak value on this curve occurs for BT = 0.685, for which we find that the maximum signal-to-noise ratio of the ideal low-pass filter is 0.84 dB below that of the true matched filter. Therefore, the " best" value for the bandwidth of the ideal low-pass filter characteristic of Fig. 1(b) is B = 0.685/T. Voy fl 10, Ja por oT Tor T re (a) laa) 10 f “BO 8 o Figure 1 04 o2 0 02 04 0685 1.0 Br Figure 2