数字信号处理-第五章
- 格式:ppt
- 大小:587.50 KB
- 文档页数:46
第五章习题与上机题5.1 已知序列12()(),0 1 , ()()()nx n a u n a x n u n u n N =<<=--,分别求它们的自相关函数,并证明二者都是偶对称的实序列。
解:111()()()()()nn mx n n r m x n x n m a u n au n m ∞∞-=-∞=-∞=-=-∑∑当0m ≥时,122()1mmnx n ma r m aaa∞-===-∑ 当0m <时,122()1m mnx n a r m aaa -∞-===-∑ 所以,12()1mx ar m a =-2 ()()()()N x n u n u n N R n =--=22210121()()()()()1,0 =1,00, =()(1)x NN n n N mn N n m N r m x n x n m Rn R n m N m N m N m m Nm N m R m N ∞∞=-∞=-∞--=-=-=-=-⎧=--<<⎪⎪⎪⎪=-≤<⎨⎪⎪⎪⎪⎩-+-∑∑∑∑其他从1()x r m 和2()x r m 的表达式可以看出二者都是偶对称的实序列。
5.2 设()e()nTx n u n -=,T 为采样间隔。
求()x n 的自相关函数()x r m 。
解:解:()()()()e()e ()nTn m T x n n r m x n x n m u n u n m ∞∞---=-∞=-∞=-=-∑∑用5.1题计算1()x r m 的相同方法可得2e()1e m Tx Tr m --=-5.3 已知12()sin(2)sin(2)s s x n A f nT B f nT ππ=+,其中12,,,A B f f 均为常数。
求()x n 的自相关函数()x r m 。
解:解:()x n 可表为)()()(n v n u n x +=的形式,其中)2sin()(11s nT f A n u π=,=)(n v 22sin(2)s A f nT π,)(),(n v n u 的周期分别为 s T f N 111=,sT f N 221=,()x n 的周期N 则是21,N N 的最小公倍数。
第五章微弱信号处理5.1 微弱信号检测技术中气体浓度检测仪中的应用微弱信号不仅意味着信号的幅度小,而且主要指被噪声淹没中的信号。
为了检测被背景噪音淹没的信号,就需要分析噪声产生的原因和规律,研究被测信号的特点、相关性以及噪声的统计特性,以寻找出从背景噪声中检测有用信号的方法。
因此,微弱信号检测技术的首要任务是提高信噪比。
它不同于一般的检测技术,它注重的不是传感器的物理模型和传感原理、相应的信号转换电路和仪表实现方法,而是如何抑制噪声和提高信噪比。
由于被测量的信号微弱,传感器的固有噪声、放大电路及测量仪器的固有噪声以及外界的干扰噪声往往比有用信号的幅度大得多,放大被测信号的过程同时也放大了噪声,而且必然还会附加一些额外的噪声,因此只靠放大是不能把微弱信号检测出来的。
只有在有效地抑制噪声的条件下增大微弱信号的幅度,才能提取出有用的信号。
为了表征噪声对信号的淹没程度,引入信噪比SNR来表示,它指的是信号的有效值S与噪音的有效值N之比。
而评价一种微弱信号检测方法的优劣,经常采用两种指标:一种是信噪改善比SNIR,它等于系统输出端的信噪比SNR和系统输入段oSNR之比。
SNIR越大,表明系统抑制噪声的能力越强。
i另一个指标是检测分辨率,指的是检测仪器指示值可以响应与分辨的最小输入值的变化值。
检测分辨率不同于检测灵敏度,后者表示的是检测系统标定曲线的斜率,定义为输出变化量y∆之比。
一般情况下,∆的输入变化量x∆与引起y灵敏度越高,分辨率越好。
但提高系统的放大倍数虽可提高灵敏度,但却不一定能提高分辨率,因为分辨率要受噪声和误差额制约。
5.1.1 本检测系统的噪声源广义的噪声是扣除被测信号真实值以后的各种测量值,可以分为两类:一是干扰;另一被称为电子噪声(狭义)。
干扰是指被非被测信号或非测量系统所引起的噪声。
从理论上讲,干扰是属于理想上可排除的噪声。
不少干扰源发出的干扰是有规律的,有些具有周期性,有些只是瞬时值。
第五章 数字滤波器的基本结构1。
用直接I 型及典范型结构实现以下系统函数21214.06.028.02.43)(-----+++=z z z z z H分析:①注意系统函数H(z)分母的 0z 项的系数应该化简为1。
②分母), 2 , 1( ••••••=-i z i 的系数取负号,即为反馈链的系数。
解:21212.03.014.01.25.1)(-----+++=z z z z z H )2.03.0(14.01.25.12121----+--++=z z z z ∵)()(1)(1z X z Y z a zb z H Nn nn Mm mn=-=∑∑=-=- ∴3.01-=a ,2.02=a5.10=b ,1.21=b ,4.02=b2。
用级联型结构实现以下系统函数)8.09.0)(5.0()14.1)(1(4)(22++-+-+=z z z z z z z H 试问一共能构成几种级联型网络。
分析:用二阶基本节的级联来表达(某些节可能是一阶的)。
解: ∏------++=k k k k k z zz z A z H 2211221111)(ααββ )8.09.01)(5.01()4.11)(1(4211211------++-+-+=z z z z z z ∴ 4=A8.0 ,9.0 , 0,5.0 1,4.1 , 0 ,1 2212211122122111-=-====-===ααααββββ由此可得:采用二阶节实现,还考虑分子分母组合成二阶(一阶)基本节的方式,则有四种实现形式.3。
给出以下系统函数的并联型实现。
)8.09.01)(5.01(6.141.158.12.5)(211321------++--++=z z z z z z z H 分析:注意并联的基本二阶节和级联的基本二阶节是不一样的,这是因为系统函数化为部分分式之和,分子的1-z 的最高阶数比分母1-z 的最高阶数要低一阶,如果分子、分母多项式的1-z 的最高阶数相同,则必然会分解出一个常数项的相加(并联)因子。
5-1 An AM broadcast transmitter is tested by feeding the RF output into a 50-Ω (dummy) load. Tone modulation is applied. The carrier frequency is 850 kHz and the FCC licensed power output is 5,000 W. The sinusoidal tone of 1,000 Hz is set for 90% modulation.(a) Evaluate the FCC power in dBk (dB above 1 kW) units.(b) Write an equation for the voltage that appears across the 50-Ω load, giving numerical values for all constants.(c) Sketch the spectrum of this voltage as it would appear on a calibrated spectrum analyzer.(d) What is the average power that is being dissipated in the dummy load? (e) What is the peak envelope power? Solution :(a) FCC power:500010lg 6.99()1000dBK ⎛⎫= ⎪⎝⎭()()()():c o s 20001000900.9501cos m m m c c b Let m t A t f H z A s t A m t tπω==→=∴Ω=+⎡⎤⎣⎦是%调制负载上通过的电压为:()707[10.9 cos (2000)]cos[2850,000]s t t t ππ=+(c)])cos[(2)707(9.0])cos[(2)707(9.0cos 707)(t t t t s m c m c c ωωωωω++-+=212500050c A =707c A V=(d)50Ω负载上的平均功率:()222A V G real 210.921505020.95000170252c A st P w<>⎡⎤==+⎢⎥⎣⎦⎡⎤=+=⎢⎥⎣⎦(e) (){}[]222PEP P 1max 500010.918050250cA m t w =+=⨯+=⎡⎤⎣⎦⨯5-2 An AM transmitter is modulated with an audio testing signal given by()120.2sin 0.5cos m t t t ωω=+, where 1500f Hz =,2f =,and 100c A =.Assume that the AM signal is fed into a 50Ω load. (a) Sketch the AM waveform. (b) What is the modulation percentage?(c) Evaluate and sketch the spectrum of the AM waveform. Solution: ()()12100(10.2sin 0.5cos )cos c a s t t t t ωωω=++2222221110.9()()12222c c cs t A A m t A ⎡⎤<>=+<>=+⎢⎥⎣⎦5-4 Assume that an AM transmitter is modulated with a video testing signal given by ()10.20.6sin m t t ω=-+,where f 1=3.57MHz. Let A c =100. (a) Sketch the AM waveform.(b) What are the percentages of positive and negative modulation? (c) Evaluate and sketch the spectrum of the AM waveform about f c .Solution:()()()()()()1110.20.6sin 3.57;1001001cos 1000.80.6sin cos m c c c a m t t f f M H zA s t m t t t tωωωω=-+====+=+()m ax m in140100%pos.m od.40%10010020%neg.m od.80%100ccc cA A b A A A A --===--===()()()()()04015c c m c m c f Sff f j f f f f f f δδδ>=------+⎡⎤⎣⎦5-5 A 50,000-W AM broadcast transmitter is being evaluated by means of a two-tonetest.The transmitteris connected to a 50-Ω load, and()1111cos cos 2m t A t A t ωω=+, where f 1=500 Hz. Assume that a perfect AMsignal is generated.(a) Evaluate the complex envelope for the AM signal in terms of A 1 and ω1. (b) Determine the value of A 1 for 90% modulation.(c) Find the values for the peak current and average current into the 50-Ω load for the 90% modulation case. Solution:(a))]2cos (cos 1[2236)](1[)(V2236)50(2000,501112t t A t m A t g A A C c cωω++=+==⇒=(b)()1111111cos 2c c os 2os []cos A t t m t A t A t ωωωω=++=to find [m (t )]min : x (θ) = cos θ +cos2θ()sin 2sin 20dx d θθθθ=--=sin 4sin cos θθθ-=125.1)5.104(5.104-=︒︒=x θ()()()111min 104.51.12min 0.905.8om t A m A x At -=→=⎡==-⎡⎤⎣⎤⎣⎦⎦m ax 1M I N 1m ax m in112236[12]2236[1 1.125]3.1250.900.57622cA A A A A A A A A =+=--==⇒=另解:(c)m ax m ax m ax m ps1112236[12(0.8)]5813.6volts116.272 50()2236[10.8(cos cos 2)]cos 0 Av c c A A I A I s t t t tfor ωωωωω=+=====++⋅=>>m ax m ax m ax m ps1112236[12(0.576)]4811.9volts96.238 50()2236[10.576(cos cos 2)]cos 0 c c A A I A s t t t tfor ωωωωω=+====++⋅=>>另解:∴ mps A 0I Av =5-12 SSB signals can be generated by the phasing method shown in Fig. 5-5a, by the filter method, of Fig. 5-5b, or by the use of Weaver ’s method [Weaver, 1956], as shown in Fig. P5-12. For Weaver ’s method (Fig. P5-12), where B is the bandwidth of m (t ),(a) Find a mathematical expression that describes the waveform out of each block on the block diagram.(b) Show that s (t) is an SSB signal.Figure P5-12 Weaver ’s method for generating SSB.Solution:)cos(212cos )(1Bt t B t V ππ=⎪⎪⎭⎫⎝⎛⎪⎭⎫ ⎝⎛=⎥⎦⎤⎢⎣⎡++-=↔==)21()21(21)()cos()()()()(313B f M B f M f V Bt t m t V t m t V π⎥⎦⎤⎢⎣⎡++--=↔==)21()21(21)()sin()()()()(424B f M B f M j f V Bt t m t V t m t V π⎪⎭⎪⎬⎫⎪⎩⎪⎨⎧<⎥⎦⎤⎢⎣⎡++-=⎪⎪⎭⎪⎪⎬⎫⎪⎪⎩⎪⎪⎨⎧<=elsewhere ,021||,)21()21(21elsewhere f ,021|| ),()(45f Bf B f M B f M B f f V f VLikewise ⎪⎭⎪⎬⎫⎪⎩⎪⎨⎧<⎥⎦⎤⎢⎣⎡++--=elsewhere ,021||,)21()21(21)(6f B f B f M B f M f V ⎥⎦⎤⎢⎣⎡⎪⎭⎫ ⎝⎛+=t B f t v t V c 212cos )()(59π⎪⎪⎭⎪⎪⎬⎫⎪⎪⎩⎪⎪⎨⎧<++⎥⎦⎤⎢⎣⎡++++-++<--⎥⎦⎤⎢⎣⎡+--+---=⎥⎦⎤⎢⎣⎡⎪⎭⎫ ⎝⎛+++⎪⎭⎫ ⎝⎛--=⇒elsewhere 21|21|,0,)2121()2121(21|21|,)2121()2121(41 212121)(559f B B f f B B f f M B B f f M B B f f B B f f M B B f f M B f f V B f f V f V c c c c c c c ccc c c c c c c c c f f B f B B f f B f f f B B f f B B f BB f f B B B f f -<<--⇒<+++<<⇒++<<-+⇒<--<-⇒<--21|21| Likemise 21212121 21212121|21:|AsideThus,[][]⎪⎪⎭⎪⎪⎬⎫⎪⎪⎩⎪⎪⎨⎧-<<--+++++<<-+--=elsewhere ,0,)()(41,)()(41)(9f f f B f B f f M f f M B f f f f f M B f f M f V c c c c c c c cLikewise⎪⎪⎪⎭⎪⎪⎪⎬⎫⎪⎪⎪⎩⎪⎪⎪⎨⎧<++⎥⎦⎤⎢⎣⎡+++--++<--⎥⎦⎤⎢⎣⎡+--+----=⎥⎦⎤⎢⎣⎡⎪⎭⎫ ⎝⎛+++⎪⎭⎫ ⎝⎛---=elsewhere 21|21|,0,)2121()2121(4121|21|,)2121()2121(41212121)(6610f BB f f B B f f M B B f f M B B f f B B f f M B B f f M B f f V B f f V j f V c c c c c c c c[][]⎪⎪⎭⎪⎪⎬⎫⎪⎪⎩⎪⎪⎨⎧-<<--++-++<<-+---⇒elsewhere ,0,)()(41,)()(41)(10f f f B f B f f M f f M B f f f f f M B f f M f V c c c c c c c cputout=s(t) = v 9(t)+v 10(t ))()()(109f V f V f s +=⇒5-13 An SSB-AM transmitter is modulated with a sinusoid ()15cos m t t ω=, where1112,500,and 1c f f Hz A ωπ===. (a) Evaluate ()ˆmt . (b) Find the expression for a lower SSB signal.cc c f c -S(t) is a USSB signal⇒ USSBf f f B f f f M B f f f f f M f S c c c c c c =⎪⎪⎭⎪⎪⎬⎫⎪⎪⎩⎪⎪⎨⎧-<<--++<<-=⇒elsewhere ,0),(21),(21)((c) Find the rms value of the SSB signal. (d) Find the peak value of the SSB signal.(e) Find the n ormalized average power of the SSB signal. (e) Find the normalized PEP of the SSB signal. Solution:()()1ˆ5sin a mt t ω= ()()()()()111ˆcos sin 5cos cos 5sin sin 5cos c c c c c b s t m t t mt t t t t t tωωωωωωωω=+=+=-()()15cos c s t t ωω=-()()22552rms c st V <>=→==()5P d V V =()()2252e st w <>=()252P E P f P w =5-18 A phasing-type SSB-AM detector is shown in Fig. P5-18. This circuit is attached to the IF out-put of a conventional superheterodyne receiver to provide SSB reception.(a) Determine whether the detector is sensitive to LSSB or USSB signals. How would the detector be changed to receive SSB signals with the opposite type of sidebands?(b) Assume that the signal at point A is a USSB signal with f c = 455 kHz. Find the mathematical expressions for the signals at points B through I . (c) Repeat part (b) for the case of an LSSB-AM signal at point A .(d) Discuss the IF and LP filter requirements if the SSB signal at point A has a3-kHz bandwidth.Figure P5-18Solution: (a)ˆ()()cos ()sin ''''A c c s t m t t m t t U SSBL SSBωω=-→+→()cos D c s t t ω=2ˆ()()()()cos ()sin cos ˆ()()(1cos 2)sin 222B A D c c c c c s t s t s t m t t mt t t m t m t t tωωωωω===+()()2c m t s t =()sin E c s t t ω=2()()()ˆ()sin cos ()sin ˆ()()sin 2(1cos 2)22F A E c c c c c s t s t s t m t t t m t t m t m t t t ωωωωω===-ˆ()()2G m t s t =()()()2H m t s t =-I :()()()()()()22,(),I c H m t m t s t s t s t U SSB m t LSSB=-=+⎧=⎨⎩To receive USSB signals, Additive V H (t ) from V C (t) at the summer.CHI(b) see part (a.) (c) see part (a.)(d) IF should be cantered atf c ±1.5 kH z,LSSBU SSBhave 3kHz BW and as small aroll-off factor as is economically feasible. LPF should have 3 kHz BW and as small a roll-off factor as is feasible, also.5-20 A modulated signal is described by the equations (t )=10 cos[(2π⨯108)t+10 cos (2π⨯103t )] Find each of the following: (a) Percentage of AM.(b) Normalized power of the modulated signal. (c) Maximum phase deviation. (d) Maximum frequency deviation. Solution: (a) ma xm i nc A AA ==0% A M∴ (b) W 502/10/2/22norm ===c A P()()()()()()310cos 2102210101502j tj t c org t e A eg t st g t wπθ===∴<>=<>=(c) m ax 10 radions θ∆=(d)kHz10102)2000(102)2000sin()2000(10)()(4===∆=∆-==πππωππθωd d F t dt t d t5-23 A MF signal has sinusoidal modulation with a frequency of f m =15KHzand modulation index of 2.0β=.(a) Find the transmission bandwidth by using carson ’s rule.(b)What percentage of the total FM signal power lies within the carson rule bandwidth?Solution:().cos 22(1)231590m m T m a m t A f tB f KHzπβ==+=⨯⨯=()()()()012320.223920.576720.352820.1289J J J J ====在卡森带宽内取了三次边频:()()()()()()()()2222222201232222201231111222222222.122222299.75%c c c c c A J A J A J A J b A J J J J ⎡⎤+⨯++⎢⎥⎣⎦∴⎡⎤=+⨯++=⎣⎦5-26 A modulated RF waveform is given by []1500cos 20cos c t t ωω+, where1112,1,100c f f K H z and f M H z ωπ===.(a) If the phase deviation constant is 100 rad/V , find the mathematical expression for the corresponding phase modulation voltage ()m t . What is the peak value and its frequency?(b) If the frequency deviation constant is 6110/rad V s ⨯⋅, find the mathematical expression for the corresponding FM voltage ()m t . What is the peak value and its frequency?(c)If the RF waveform appears across a 50Ω load, determine the average powerand the PEP.Solution:.:a PM.:b FM()22150012522s c RF P A KW ===波形的规一化功率:()12500025005050s real PEP P P W P Ω====通过50负载后:5-46 A digital baseband signal consisting of rectangular binary pulses occurring at a rate of 24 kbits/s is to be transmitted over a bandpass channel.(a) Evaluate the magnitude spectrum for OOK signaling that is keyed by a baseband digital test pattern consisting of alternating 1’s and 0’s.(b) Sketch the magnitude spectrum and indicate the value of the first null-to-null bandwidth . Assume a carrier frequency of 150 MHz.(c) For a random data pattern, find the PSD and plot the result. Compare this1()cos[()]500cos[20cos ]100()20cos 210001()cos 210005c c c P S t A t D m t t t m t tm t tωωωππ=+=+∴==[]()166()cos[()]500cos[20cos ]10()20cos 2100020()sin 2100021000104sin 21000100t c c c f t S t A t D m t dt t t m t dt t m t t tωωωπππππ-∞-∞=+=+∴==-=-⎰⎰result with that obtained in parts (a) and (b) for alternating data. Solution:(a) Evaluate the magnitude spectrum for OOK signaling that is keyed by a baseband digital test pattern consisting of alternating 1’s and 0’s.()()cos c c s t A m t t ω= , m (t )为单极性 OOK : ()()c g t A m t =0000000/4/4/4/4/2/20000()1sin(/2)2/2/2sin(/2)()()() 2/211 w here:22jn tn n T jn tT jn tc n c T T jn jn c c cn n n bg t C eA eC A edt T T jn A A e en T jn T n A n G f C f nf f nf n R f T T ωωωππωππππδδπ∞=-∞-----∞∞=-∞=-∞===⋅--==-⎡⎤=-=-⎢⎥⎣⎦===∑⎰∑∑*1()[()()]2c c S f G f f G f f =-+--(b) Sketch the magnitude spectrum and indicate the value of the first null-to-null bandwidth. Assume a carrier frequency of 150 MHz.48null B K H z =(c) For a random data pattern, find the PSD and plot the result. Compare this result with that obtained in parts (a) and (b) for alternating data.48null B K H z =The null-to-null bandwidth is the same for both (b) and (c). Both have sinx/x type spectral envelope.5-47 对于BPSK 调制,重做5.46的(a)(b)(c)()()U nipolar N R Zc g t A m t =→22sin ()()4cb g b b A fT P f f T fT πδπ⎡⎤⎛⎫⎢⎥=+ ⎪⎢⎥⎝⎭⎣⎦1()[()()]4s g g c c P f P f f P f f =-+--()()cos c c s t A m t t ω= ,(a)(b)()/2/2b b b b t T t T p t T T ⎡⎤⎡⎤+-=-⎢⎥⎢⎥⎣⎦⎣⎦∏∏ ()()()0nm t p t t nT δ=*-∑()()()()()2/22/22sin b b bbj fT j fT b b b b j fT j fT b b b b bP f T Sa fT eT Sa fT eT Sa fT e ej T Sa fT fT πππππππππ--=-⎡⎤=-⎣⎦=()()()()12sin()22sin()2/2sin(/2)2b b b nb b b b n b n b n Mf j T Sa fT fT f T T n jSa fT fT f T n j Sa n n f T ππδππδππδ⎛⎫=⋅- ⎪⎝⎭⎛⎫=- ⎪⎝⎭⎛⎫=- ⎪⎝⎭∑∑∑()()[()()]2c c c A S f Mf f f f f δδ=*-++(c) ()()cos c c s t A m t t ω= m(t)为极性NRZ()()c g t A m t =()()22sin b gc b b fT P f A T fT ππ⎛⎫= ⎪⎝⎭()()()14sgc g c P fP f f P f f ⎡⎤=-+--⎣⎦5-49 Evaluate the magnitude spectrum for an FSK signal with alternating 1 and 0 data. Assume that the mark frequency is 50 kHz, the space frequency is 55 kHz, and the bit rate is 2,400 bits/s. Find the first null-to-null bandwidth. Solution:The result if given by (5-86).f 1: : =50KHz f 2 : =55KHz R : = 2.4 Kbit/s where f2>f121f f h R-=A C := 1122c f f f +=Solution:By problem 5.46:0000*sin(/2)()()() 2/211w here:221()[()()]2c n bc c A n G f C f nf f nf n R f T T S f G f f G f f πδδπ∞∞-∞-∞⎡⎤=-=-⎢⎥⎣⎦====-+--∑∑()()1122()cos()cos()FSK n b c n n b c n s t a p t nT A t a p t nT A t ωθωθ⎡⎤=-⋅+⎢⎥⎣⎦⎡⎤+-⋅+⎢⎥⎣⎦∑∑()()nbnm t a p t nT =-∑null (5550)2 2.49.8kH zB =-+⨯=b.()()c g t A m t =22sin ()()4c b g b b A fT P f f T fT πδπ⎡⎤⎛⎫⎢⎥=+ ⎪⎢⎥⎝⎭⎣⎦1()[()()]4s g g c c P f P f f P f f =-+--null (5550)2 2.49.8kH zB =-+⨯=5.53 A binary baseband signal is pass through a raised cosine-rolloff filter with a 50% rolloff factor and is then modulated onto a carrier. The data rate is 64 kbits/s. Evaluate(a) The absolute bandwidth of a resulting OOK signal.(b) The approximate bandwidth of a resulting FSK s ignal when the mark freqwency is 150KHz and the space freqwency is 155KHz.(Note: It is interesting to compare these bandwidths with those obtained in probs. 5-46 and 5-49) Solution:()0643222R f K H z ===()()010.5321.548B f K H z =+=⨯= a. OOK()()0210.548296OOK B f KHz =+=⨯=b. FSK()()()212155150248101FSK B f f B KHz =-+=-+⨯=5-58 Assume that a QPSK signal is used to send data at a rate of 30 Mbits/s over asatellite transponder. The transponder has a bandwidth of 24 MHz.(a) If the satellite signal is equalized to have an equivalent raised cosine filter characteristic, what is the rolloff factor r required?(b) Could a rolloff factor r be found so that a 50-Mbit/s data rate could be supported?Solution:for QPSK M=4, l=2(a) If the satellite signal is equalized to have an equivalent raised cosine filtercharacteristic, what is the rolloff factor r required?0.6 6.130)24(2)1(230)1(24===+⇒+=⇒r or r r(b) Could a rolloff factor r be found so that a 50-Mbit/s data rate could besupported?()r f B +=10()021T B f r =+()021T B f r =+502424(1)(1)225r r ⇒=+⇒+=A rolloff factor ,r could not be found support 50Mb/s QPSK signaling5.62 Assume that a telephone line channel is equalized to allow bandpass data transmission over a frequency range of 400 to 3100Hz so that the available channel bandwidth is 2700Hz and the midchannel frequency is 1750Hz. Design a 16-symbol QAM signaling scheme that will allow a data rate of 9600 bits/s to be transferred over the channel. In you design, choose an appropriate rolloff factor r and indicate the absolute and 6-dB QAM signal bandwidth. Discuss why you selected the particular value of r that you used. 5.62 Solution:31004002700Q AM B H z =-= 49600162424004M l D Bd ==→=→==()()131004002400127000.125D r r r +=-→+=→=62400dB B D Hz ==习题:5.69 5.715.69 Plot the MSK Type I modulation waveforms x(t) and y(t).5.71 Plot the MSK TypeⅡmodulation waveforms x(t) and y(t).5.67 For π/4 QPSK signal,(a) Caculate the carrier phase shifts when the input data stream is 10110100101010 , where the leftmost bits are first applied to the transmitter.(b) Find the absolute bandwidth of the signal if r=0.5 raised cosine-rolloff filtering is used and the data rate is 1.5Mbits/s.Solution:For π/4 QPSK signal, M=4 ,l=2 0 1.52(1)(1)(1)(10.5) 1.1322T R B f r D r r M H z =+=+=+=+=。
第5章 数字滤波器的基本结构5.1 学习要求1 掌握IIR 数字滤波器的基本网络结构,包括直接型、级联型和并联型;2 掌握FIR 数字滤波器的基本网络结构,包括直接型、级联型和频率抽样型;3 了解数字信号处理中的量化效应和数字信号处理的实现。
5.2 学习要点5.2.1 数字滤波器的结构特点与表示方法一个数字滤波器可以用系数函数表示为:01()()()1Mkk k N kk k b zY z H z X z a z -=-===-∑∑ (5-1) 直接由此式可得出表示输入输出关系的常系数线性差分方程为:1()()()N Mk k k k y n a y n k b x n k ===-+-∑∑ (5-2)由式(5-2)看出,实现一个数字滤波器需要几种基本的运算单元—加法器、单位延时和常数乘法器。
这些基本的单元可以有两种表示法:方框图法和信号流图法,如图5-1所示。
用方框图表示较明显直观,用流图表示则更加简单方便。
z ⊕aa单位延时乘常数相加方框图表示法信号流图表示法图5-1 基本运算过程的表示5.2.2 无限长单位脉冲响应(IIR)滤波器的基本结构无限长单位脉冲响应(IIR)滤波器有以下几个特点:(1) 系统的单位脉冲响应()h n 是无限长的;(2) 系统函数()H z 在有限z 平面(0z <<∞)上有极点存在; (3) 结构上存在着输出到输入的反馈,也就是结构上是递归型。
同一种系统函数()H z 的基本网络结构有直接I 型、直接Ⅱ型、级联型和并联型四种。
1直接I 型直接型按式(5-2)差分方程式将输入采样值(序列))(n x 延迟并乘以系数k b ,将输出采样(序列))(n y 延迟并乘以系数k a ,再把它们加起来,这种结构称为直接I 型,结构流图如图5-2所示。
由图可看出,总的网络)(z H 由Mkk k b z-=∑和11Nkk k a z-=-∑两部分网络级联组成,第一个网络实现零点,第二个网络实现极点,从图中又可看出,直接I 型结构需要N M +级延时单元。