SIGNALS & SYSTEMS Uday Kumar S. Published by : Prism Books Pvt. Ltd # 1865, 32nd Cross, 10th Main, BSK II Siage Bengaluru-560 070. Phone: 080-267 14108 Telefax : 080-26713979, e-mail : info@prismbooks.com www prismbooks.com Also at Chennai; Tel ; 044-24311244, e-mail : prismchennai@prismbooks.com Hyderabad: Tel : 040-27612998, e-mail : prismhyderabed@prismbooks.com Kochi; Tel: 0484-4000945, e-mail : prismkochi@prismbooks.com Kolkata; Tel : 033-24297957, e-mail ; prismkolkata@prismbooks.com © Prism Books Pvt. Ltd. Fifth Edition: 2009-10 Reprint 2018, 2015, 2014, 2013, 2012 Pages 4a2 Price % 395 [No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permissian in writing from the publisher. ISBN 978-81-1286-692-1 Printed at: {gh Aditya Printers { Printing & Publishing House . / National Award Winner for Excellence jie Printing. #1, Hanumanthanagar, Bangalore-19 eee en / Ph; 080-26606776, Fax: 080-26606566 ~~ mail: prakash(@adityaprinters com, web: wwwadityaprinters.comChapter 1. 2. 4, Contents Page INTRODUCTION 1.1 Definitions of a Signal 1 1.2 Definitions of a System 3 1.3. Blementary Continuous-time Signals 5 1A Elementary Diserete-time Signals 10 1.5 Basic Operations on Signals 15 16 Classification of Signals 35 1.7 Syste viewed as Interconnection of Operations 76 L8 Properties of Systems 78 1.9 Exercises 96 ‘TIME-DOMAL REPRESENTATIONS FOR LTI SYSTEMS 2.1 Intreduetion wT 2.2 Impulse Response Representations for LTT Systems 107 2.3 Properties of Impulse Response Representations for LTT Systems 140 24 Differential | Difference Equation Representations for LTT Systems 162 2.5 Block Diagram Representations 184 2.6 Exercises 198 FOURIER REPRESENTATIONS OF SIGNALS 3.1 Introduction ‘200 3.2. Fourier Representations for Signal Classes 200 3.3. Orthogonality of Complex Sinusoidal Signals 201 3.4 Disorete-time Periodic Signals : The DTFS 202 3.5 Continuous-time Periodic Signals : The FS 221 3.8 Discrete-time Non-periodie Signals: The DTFT 24a 3.7 Continuous-time Non-periodic Signals : The FT 280 3.8 Exereises 318 APPLICATIONS OF FOURIER REPRESENTATIONS 4.1 Introduction 324 4.2 Frequency Response of LTH Systems 824 4.3 Solution of Differential | Difference Equations 830 4.4. Fourier Transform Representations for Periodic Signals 3354.5 Sampling of Continuous-time Signals 4.6 Exercises 5. THE Z-TRANSFORM Bl Introduetion 349 5.2 The Z-Transform 349 5.3. Properties of the Region of Convergence 362 5.4 Properties of the Z-Transform 369 5.5 Inverse Z-Transform 982 5.6 Transform Analysis of LTI Systems 400 5.6.1 Transfer Function and Impulso Response 400 5.6.2. Relationship between Transfer Function and Difference Equations 403 5.6.3 Stability and Causality 406 5.6.4 Inverse Systems 409 5.7 The Unilateral Z-Transform 4 5.7.1 Solving Difference Equations 413 5.8 Exercises 420 Appendix-| DTFT of Some Basic Signals 426 Appendix-II_ Properties of DTFT 427 Appendix-III_FT of Some Basic Signals 428 ‘Appendix-IV_ Properties of FT 429 Appendix-V _ Z-transform of Some Basic Signals 430 Appendix-VI_ Properties of Z-transform 431 INDEX 432Chapter 1 mm Introduction 1.1, DEFINITIONS OF A SIGNAL A signal can be defined as 4 funenon that conveys mformaron Although signals an be represented 1 many ways, un all eases the information # contasned sn seine atically as a function of one pattern) oF variauions. Signals are represented anathe or more dependent variables IY che-Tanetion depends on a single variable, she signal ws said co be ome divensional. Eg., Speceh signal, A speech signal is represented athemaneally as 2 function of time wherein the amplitude varies sith time depending on che spoken word and the person vho speaks at If the timetion depends on own oF more variables, the signal is said to be muliidimensional, Eg. (mage signal A photographie image is represented ay 4 brightness finetion sf ypatia variables Usually the mdependent vanable uf the mathematical representation of «signals taken as time, Buc in some specific eases the independent variable may not be nme Real-Life examples for signals tn his section, we have dicted few exatuplen for seal signals 1) By hstening © the heart-hear of 4 patient a doctor is able tw diag presence or absence of disease, The quantity (heart-beat) represents signal that convey mformiation 1 the doctor about the state of healdh of the patient es regarding. hy) bn hstemng to 2 weather Forecast over the radio, we wet quant vanstions im temperature, humidity, the speed of wind ete The signals represented by these quantities help as to decide whether v0 go vu five a walk Hasieally there are qo types of signals. (2) Connnucis-cime signa ane an) Discrete come signal ( Continuous-time signal A signal x(t} 1s said to be + conumos-tine sual at it hay value ool amphtade for all ume ‘t (ie, the independent variable “t' eontinuous). Fig. || represents an example of a conmnucus-time signal whose amplitude varies continuously with time Continmous-time signals arise naturally when a phystcal phenomenon (eg,, heart-beat acoustic pressure variation ete.) 1s converted suite an electrical siginal using appropriate rransducer2 Signals & Systems x 5] Fig. 1.4. A contiwous-tine signal iscrete-time signal A diverete-time squat is defined only at diserete instants of time. (i... the independent variable has diserete values only which are usually waiformly spaced) A discrete-time signals are represented mathematically as sequence of numbers. A sequence of numbers x in which the n number in de sequence is denoted by x(n) is writen as, x= (aln)} ocr where nis an inter. Practically such sequences are usually obtained from a continuous ~ time signal by sampling ie ata uniform rate, Consider‘ the sampling period and 'n’ denoze an integer (-22 v9) fa represent the input-output relation of a continuows time syste ref-tinne quton is one where discrete-time input signals are applied which results in discrete-time output signals. It ean be represented pictorially as shown inSais & Systems Discrete-time System | fs va Fig. 1.4 raresontaton for discrete tm system where sin} 1 dhe input signal and y(n) rs dhe cutpur signal Alternauvely, we ean represent the input and output relanan ata uscrete-tine syste by she notation as writren helow etn) > ta Examples tor systens lagatty oF me Yorwe Computer ie Fig. 1.5 Automatic speaker recognition system ae oe ree ee = : : fig. 1.6 Communication system Real-Life Examples for Systems iy has section, ve will discus sone of the realehfe systems i bret 1, Communication Systems The three basie elements of conmunieation system are (1) Transiaer (11) Channel (it) Receiver as shown in Fig. 16. The tarsmniter and mecrer are placed at a distance apart ‘which are connected by a physical meditam called chase! The channel may be free space pti! fiber, co-axial cable etc, The aansnmiter convents dhe message sagnal (eg. speech Signal, video yignal ete.) produced by a source of information into a fora sutable fir Gansnnsson over the chatinel When the wransmutted signal travels through the chanel stwonld he distorted due to the physical characteristics oF the channel, Tu adklitton wo chi noise and interfering signals originating from other sources contaminate the message signals. The receiver receives the distoreed signal co reconstruct 1 into 4 recognizable forny or into an estimated form of the ongnal message signal Thus. the receiver does the reverse process chat of cransmtter and als reverses the limination. amplification of weak signals te cect af she channel such as nox 2. Control Systems Fig. 1.7 shows che block dugram representation of 4 eloséel-toop control systems ot Feedback control systemoduct Disturbance ey sate CZ const FS] Pen = Zo ome Sait oak ; Signal Bit) Fig. 1.7 Block ctagram of close loop contol system i any control system, the plane is represemted by snathemaueal operations that sgenerate the output y(¢) in response to the plant snput v(t) and the external distarbanes d(t), The sensors existing in the feedback loop measures the plant output y(t) and converts it into another form B(t) known as fedoack signal. It 1s compared against the reference input x(8) tw produce an erm signal e(®). This error signal is applied cw + controller which in eum produces he aemarng rghsl v(t) that performs the controlling action of the plant For example, ian arreraft landing. yystem. the plane is represented by the:sireratt becly ind actuntor, The sensors are used by the pilot to determine: che lateral pesition of the riverafeand the controller isa digital eompurer 5, Remote Sensing System Remote sensing is defined as the piovess of accjuningg satormavons about a abject at merest without being in physical coweact with 11 The acquisition of ynlarmnations accomplished by detecting and measuring the changes dat the abyect ereates on the sortenanding field. This fickd may be clectrovnagnctic, acoustic. nuagnetic en 2 The acquisition of informaions can he done 1 4 passive manner by lisienmg 6 te field that naturally ernitted by the object and processing rt ar by purposely illminating the object with a defined field and processing the echo 1.3 ELEMENTARY CONTINUOUS-TIME SIGNALS The elementary or beste signals occur fiequently in nature. These signals sre the baste bburlding blocks for constructing more compley signals. Many physical signals thar vecur n_be modelled using these elementary siginals Some of the unportnt clementny continuons-rinie signals ae 4) Exponential signals 1) Simsoidal signals st) Exponentially damped suum als ¥) Unie step fanetion ¥)— Unic impulse Hance Ai Unie ramp finetion6 Signals & Systems 4.3.1. Exponential signals ‘Areal exponential continuous-time signal is given by x)= Ce" (a 2s knows as the amplitude of the whore both ‘C° andl ‘a! real eonstant. In eqn. 1.1, exponential signal at =0. Ifa <0 (eq ‘a is negative), the signal x(¢) is known as decaying exponential signal and if a> 0 (vc. "a" is positive) then (0) is called groin exponintial signal Fig. 1.82) and (b) shows the real decaying and growing exponential signal respectively. S xy=Ce" x(=ce Fig, 1.8(a): Real decaying Fig. 1.8(b) : Roal growing ‘xporontial signal (ie, a 01 In eqn. 1:1, if (C? or 'e of both are complex sumbers, then (1) is known as comtinewus-tine cornplex exponential signal Note; Ineqn. 1.1, consider C=1 and ‘a? is imaginary, ic, x(t) =e (1.2) 25 The signal «(:) given by eqn. 1.2 is periodic with fundamental period T = A continuous-time signal 4(0) is periodic with period T, if xQ)=8G+7) =) i) CFT) (a) For x(t) to be periodic ett we died cet = eit, Eqn. 1.3 is valid only if ote (6) If @=0 then &”Intioduotion 7 IF, #0 then the fundamental perind of x(t) is the smallest valie of T and is sven by 07) Substituting equ. 1.7 ineqn, 1.6 we get = OM = coset jsinde= 1 as) Therefore ost and e:%! are periodic with fundamental period TH 1.3.2 Sinusoidal signals A sinusoidal signal is given by, XQ) = Acos (@,0+6) ones AES where =2nf, = angular frequency (rad /see) = linear frequency (12) 4 = phase shift (radians) ‘The fandamental period of the signal x(t) in eqn. 19 is T=2% (sec.) "1.3.3. Exponential damped sinusoidal signals ‘An exponentially damped sinusoidal signal is given by, aro (9 =e" sinwe whet As increases, the amplitude of sinuscidal oscillation decreases exponentially and approaches zero as t-» 0, An exponentially damped sinusoidal signal is shown in Fig, 19. xt) Fig. 1.9 Anexponentially damped sinuscidal signal 1.3.4 Unit step function : u(t) The continuous-time unit step function is defined as, eo itso ult)8 Signals & Systeme Teas shower ae Big 110 ae H Fig. 1.40 An unt stap function 1.3.5 Unit impulse function : 5(¢) The commnuous-tmic anit nipulse fuetna §(¢) 1s defined This ficou 8{t) walso known as Dire dela funenan Phe contnucnis-tne mit anipalse Ranetnon 6¢t) a8 show a Fre. | UT 80 0 Fig. 1.41 An anit empuse tuoctin Fan 110 mdicates chat the area covered by an amit impulse fimenon is mary The smpulse finction 8(0) is the derwative of the step fuanenon with respect to time Consider « non-dealized unse step function «ay(®) as shown an Fa, F122) The ierwarive of us(0) with respect ra time *€ shown in Fi. E12Ub) ayn) 2) = le) an ana Fig 1.198) Fig. 119(6)Er Intoduction ° 8c = A (any The ars of oncsion Be) = A-f= sta Fig M12b).008 0 Perc aoget aligned aeoinelo anki w=, 4, 40 ug Therefore unit impulse fanesion has, (Zero width (ii) Infisite heighe (ii) Univ area or unit strength A graphical representation of impulse finetion with strength ‘Kis shown in Fig. 1.13. The strength indicates the area under the impulse. #8(0) ces 0 Fig. 1.13 Impulse function with strongth Properties of continuous-time Impulse function : “ foro d= (i) fka@ =k Gi) (980) = x10) 8(0, co) feo Ho dk = x10) () x1). 8(t+4) = a(t) 6(leK) => (sampling property of impitlse function) oi fro sees (6) => (sifting property of impulse function)‘0 Signais § Systeme 1.3.6 Unit ramp function 7(t) en Dred A ramp finetion is defined 26 r( Its the mcegral of the ume step Function u(t), An anit ramp Fanconi 9 sbi Fg 114 He) ‘ent slope ° Fig. 1.14 An uit ramp function 1.4 ELEMENTARY DISCRETE-TIME SIGNALS The elementary discrete-time signals are direct counterparts of the eoummuous-tune signals described in section 1.3. Many of the characteristics of basic discrete-time signals are directly analogons to that of basic continuous-time signals, But. few important characteristics differ in case of discrete-time signal from that of continuous-time signal Some of the important basic discrete-time signals are Gi) Exponential signals Gi) Sinusoidal signals (iii) Exponentially damped stnuontal sgital (iv) Unie step sequence (9) Unie impulse sequence (vi) Unie ramp sequence 1.4.1. Discrete-time Exponential Signals A real exponential discrete-time signal or sequence is gven by, x(n) = Ca" ay where ay and C, and B are real constants, In eqn. 113, (C's known as the amplitude of the sequence at n = 0. If || <1, the signal decays exponentially, Furthermore, if a<0 (j.t..0 's negetive), then the sign of a(n) alternates ic., wien ‘a’ is positive. x(n) has positive value and when ‘n' is negative, x(n) has negative value. The plots of x(n) = Cat for a> 1, O! x(n) € te . Sea etes @ () 0 Gsfemg property of rmpulse sequence) nz0 =0 nso u(n) = dn) + 8-1) + (m-2) + + (nen) Note: wn) ti) = SF sty anak ote) esIntrocuenor 1.4.6 Discrete-time unit ramp sequence : r(n) A discrete-time ramp sequence is defined as x(n) 220 n 1. the signal y(t) is a compressed version of x(t) and if 04 < 1 the signal y(0 18am expanded version of (0 Eg. Lot x(0) sas shown in Fig. 1.20 taking the ro) Fig. 1.90 A continseus-tie signal x(t) The signals x(2¢) and x(0) are shown in Fig, 1.21 and Fig, 1.22 respectively, ey LL $a Fig. 1.91 Compressed version of x(t) bya factor 2wwe) [ . a a Fig. 1.92 Exparded version of x(x) by a factor 2 In the discrete-time sequence vin) =a(kmy RO (135) where “kx an integer. W le> 1, seme samples ot xin) wold be los Eg. Let x(n) isas shown in Fig 1.23 xo ie 2 1s og eet TL || ao a4 + aS 2 12 Fig. 1.23. discrete-time sigral v1 Phe signal x(2n is shown below in Faz. 1.24 acany a or Fig. 4.94 Comprossod werion of x(n) by a factor b) Time Shifting = Let x(0) be a conoions-nnn sensal Then che signal yO) = x(t (1.36) 1s knowsas tine shified version of xi. where (1s the ume shite If 6,20, the waveform of the signal is shifted to the right HV t,<0D the waveform is shifted to the let. Lett) is shown in Fig, 1.258 ‘Sionals & Systems Fig. 1.95. continuous-time signal Then, x(t-2) and (¢+3) is shown in Fig, (.26 and 1.27 respectively Fig. 1.27 Alef shifted (advanced! signal 6) Reflection : Let x(¢) be a continuous-time sagnal. Then the signal y(0) = C9) 18 known as the reflected version of x(@) about the amplitude axis Example for a continuous-time xt) and its reflected! version x(-) shown in Fig, 1.28 and 1.29 respectively. 2 Fig. 4.98 A continyous-tma signal x(t)Iniroduction Fig. 1.29 Reflected version of (1) Precedence Rule : Let the signals x(2) and y(t) are related by the following eqn. [.37, 0) = ata) (1.37) | Tor yet y(t) from x(t). we have to perform both tme shifting and time sealing operations, | Pur t=O in eqn. 1.37 we get | yd) = x(-b) 1.38) Par r= tya my eqn. 1.37 we get 1.39) | (ba) = 900) Onee we obtain y(t) from x(@) by perfomung ome shifting and one seahng ‘operation, it unust satisfy the eqn, 1.38 and 1.39. This » possible only if the ome shifting operation 1s performed First on x(1) whneh yields am internediate signal v(t) given by. veo = ateb) 1140) Newt the time scaling operation is performed on v(t) to obtain v(t) ves y(t) = oat) = eb) vat ACS ee TR EE Example 1.1 Sketch the following signal x(0) = a(t) = 9(t-2) Solution: Given x(Q) = u(t) = u(t-2) The signal anne step fianetion nt) shew m Fig. PATA my, fig. Pista20 Signals & Systeme The signal a(t-2) w obtained 6 shutting qe) we rhe rghe by Sanit + shown Fig PLL? x(t) = 0-0 xit) = 1-0 xi) = 1-190 The signal x(t) is shown in Fig, P1.1.3 xi) Example 1.2. Sketch the signal.x(t) = -u(t+3) +2u(C+1) - 2u(t-l) + a(-3) Solution : Firstly, we sketch the signals x(() = -u(e#3) 5 =2(0 = 2u(e#1); xa() = -2u(tel) and. x4(0) = u(-3) as shown in Fig. P12 (a) to (4) x(t) FThc 2 (b) 3 Sse Le Se, bso 19) eed ops apcaan [a= [aps BE ok he eo ee Oa eae oe asin be) 1 Oa a aw Te : Fig. 91.2 The signal s(t) = x(t) +22(0)+2,(0-+Xu(0 is obtained as below and plotted in ig, PLC). Fort<-3 x(t) =arlt) =) =A = 0 ‘ +0+040=0 Ford3 1.3 Sketch the signal, 2) = 1041) (9 +(e) Firstly, we draw the signals,Ea wi) = es 0 (by te a s(t) Fig. PA a Signals & Systems ar(t) and agit) = r(@1) as shown un Fig, PLB and (¢) respectively and x(t) 1s shown int Fig. PL3fel) Example 14 A continuous-time signal x(t) shown in Fig, PL¢. Draw the signal ¥(0) = [N)+x(2-0) fed “) 4 Opt Fig. Pa,” © (a) we Ey axe) x40 a 2 ) 1 2 a ‘ wo 0] 1 2 a . j Ls treo rat-g 2 Of 1 2 3 :2 Signals & Systems yy =3 j0stsi =0 ‘otherwise Example 1.8 A continuous-time signal x(0) and g(t) is shown in Fig PLS. and P1S.2 respectively Express x(t) in terms of g(t) Fig. P 1.5.1 Fig. P 1.89 Fig. P1.5.3 Fig. P54 Fig P55Introduction 2 aC} Se eet i Fig. P1.5.6 Adding g(t}. g(48) and g(t) yields the signal s(t) as shown in Fig, P1.5.6. x0) = a) + 80 + eG0 Example 1.6 For the continuous-time signal x(0) shown in Pig, P1.6, sketeb the signal y@) = Be+2) ,[o Z a Fig. P16 t 7 ° 3 Solution: According e9 precedence rule, first we obtain an intermediate signal v(t) representing time shifted version of x(t) and then compress ity & factor of 3 wo ger (= vi3t) = xGer2) v(Q= x{t+2) | 1.64 yl= vAn= xe#2) Fig. 71.6.226 Signa & Systems Example £7. Porte contmoous-tme signa s(Q sbiowaran Pre: PL 0 tprestaus esaple) pobnin ye) = 9130) + v2), Solution; The siggrals x(t) and x(3e#2) are shown ay Fig, PLT. and Fay PL72 respectively, ae Fig. PA.7.2 yi)= 00+ 342) a zy A dow lek 4 2 3 Fig. PIL3 Wehave y(t) = 813) + 60142) For ec t<-1 .xGQ=0 and xt? For-lete x20 yi Forces Ys 2G?) vit) Fore Ye rQD=0 and v3 yo Example £8 For the coutsous-time sinil s( shown m Fig, PLO, draw yie) = 2-2) Soluion: — y(v) = a(2(©2)) y() = s2e4) Accordingee the precedence rule, first let us obtam she uiteviediate sigual v() which 4) asshown in Fag, PLB | yi 0 (3e+2) 1s ane shafted version v(@) fie. v(snwroduction 0 ” y= ated) 1 gh foe eae v(t) = 920 = x(t asshowwn mn Fy, PLB YOO =V2D= M2 =UALA Fg P1839 ets? %s @ 6 Example 1.9 For the continmoay time signalxtt) shown m Fig, P1.6. draw y= Solution: We have to find ye) = 8-261) In this case, we have to perforin 3 operations the following order (i) Time shiizing (i) Compression & (ii) Time reversal Time shifted version ¥(0) tte 1) is shown m Fig. PL9.1 Fig. 71.9.1 The compressed sygnal 203 (29 = rte) rs shown an Fig, P92 ett) =v(20) Fig. P1.9.9The rime ceversal version y(t) Signals & Systems = Het) = (201) yt) =r) 9-2-1) / \ ig. P1.9.3 fo Example 1.0. Two continuous ome signals x(Q) and y() are given in Fig, PHOT and P1.10.2 respectively. Draw 2) = <2). y(2e41) =) fig. P1.10.2 We have co draw 2(0) = «(29 y(2e+ 1) and aris shown on Fag. PE LOS x20 ig, P110329 yarn ty a Fig. 91.104 Q)=s(2).y2r4 1) Fig. 1.105 fered sane & asa 0-0 ordered sain 6 yarenen ao=0 por be red sx@bet) & yer bat sine Ford 3 M2920 & vi2t+)=0 Example 1.11. For the x(Q and y(Q shown m Fig, PH.10.1 6 P1102. draw 2( Solution: Wehuve ro deaw 2(0) = (0). vel. ve. 208) = x0). CED) F : Fig. P4414se ‘Signais & Systems nyeand retleernon operation yt) as visi) abtansed by pertormmg anne shih clussonv it Fig PLT 2and Fag, PILL 3 respertovely vineMe}) ao 1 Oo 1 7s Fig. Pate At) = v0 = ve I) Fig. PLAT The stenal et) = he) EH =a(e| yf Hees shower an Fig PENTA aN yl) Fig. Pata Hesample 1.12 Che discrew-rime signals y(ny and y(o) are shown ny Fry, PL 12.1 and Fie PY 122 respectively. Sketel the signal (i) = 9(2n) vindon ” sto : | [ee] | ee tai aeegiied supe fig Phas von 4-28, fe Fig. P1129 olution : The igual x(2n) and y(n) are shown m Fig, PL 12 Rand Pug Pl /2 4 respeervely v(2u Jo aes Fig. P1193 vind) Fig. PI19.4 The signal (0) =6(2n}. ylnrd) ws shown im Fig. PL 125(2). y(n-4) Fig. P1195 From Big, PI 12S and 1.124 we have s2my(n-d) =O 0 Forn<0sy(n-4)=0) Forn=0:x(2n)=0 ein) = 2(0) For (2h 1c atu) = 2(1) Forn> 1.x) ain) = 0 a & yin) Example 143 Adiscrete-time signal x(n) s shenwn in Fig, PES. Sketch the signa yon) = x(a) un) so) Fig. P1.13 is shown in Fig, PLLA Solution: The signal w(2- 42-0) Fig. P1.13.1Now, tne signal y(n) = x(n). u(2-n) i shown in Fig, PLS. y(0) = s(n)-a(2-n) 1 S3ss7 ort s Fig. P1.13.2 . a(n) = h(n} . 8(n-1) ha Fig. P1.14 Solution; Given: x(n) = hGn) d(n-1) The signal h(n) is shown in Fig. P1.14.1 below hn) Fig, PI14Signals & Systeme Phe gual Ste 1) 4s shown an Fig. P1142 below Bln!) a ? Fig. 1.14.2 He gal xin) ha) 8-4) is shown in Fig. PL14.3 below x(n) ie te Tes? 8 ig, PIAA EEE Example 1.15 continuous time signal is shown in Fig, PLLIS Find y(t) = xGt) [8(+4%) - 372)] xo) 2 y 1 oy te 78 : foe Fig. 1.15 Solution; The plocof BE#Y:) - 3% ) is shown in Fig, PLL15.1 below: BH) Fig. P1151w(0=28( -) 1 oft) Bitty = 94) B(HK | Fig. P1.15.9 "1.6 CLASSIFICATION OF SIGNALS The different methods of classifying sigmals bused nn diferent Feararey a (i) Continuous-time and Diserete-tinne siunals (i) Even and Odd signals (ii) Perindic and Non-periodic signals (iv) Deterministicand Random signals (v9) Energy and Power signals 4,6.1 Continuous-time and Discrete-tme signals, The difference between continuous-time signal and discrete-time seal ts alread | explained in section 1.1 k "16.2 Even and Odd signals A continuous-time signal x(t) is said to be an even signal af 1 satistics the condition (40) =2x{¢) forall‘? and itis said co bean odd signal ft satisfies te condision.x(-t) = -x{e) for all's Similarly, « discrete-time signal x(n) 48 said to be an even signal if it sansties the condition x(-n) = a(n) forall’ and it is said to be un odd signal if it satisfies the condition x(-n) = x(n) forall Even signals are symmetrical about the vertical axis or the time origin whereas odd signals are antisymmetric about the vertical axis or the time origin, The above definitions of even and add signsl holds good for ony teal vatued signals But complex valuee signals are characterized by conjugate symmetry. A complex valued signal x(1) is suid to he conugaie symmetric if it satisfies the condition, xe) =) sae (1:42) where ‘* denotes complex conjugation.* Signals & Systems Let a = xnlt) + Jn) where aiy(t) the real pare of (0), ay(0) is the imaginary part of (0) and Then complex conjugate of x(2) is. ©) = xn -1 10 oC) Decomposition of a signal Accontinuousetime signal a(t) can be decomposed into @ sum of two signals, one of which is ever (2) the other is odd ,() such that, 7 a) = 50 #200) (145) Fors,(t) to be even, aC) = 50 o (146) and x,(0) 0 be odd, acd = 60 an tm ot imeqn, 143 we get (2) +569 AC) = KY (48) Solving for x,(0) and x,(t) from eqn. 1.45 and 1.48. we get, x0 = fixcotxC0] 49) 1 & = Fl) - x0] (1.50) tne signal.x(n) can be decomposed into surn oF ewo signals, one nd the other is odd x(n) such that, Similarly.a disere of which is even. x(a) = x(n) + (1) (51) where ain) =} [x(n +n) (152) and x,(0) =F pn) -s6-m)} (1.53) LS Examplet.t6 Find the ever and odd components ofthe signal x(t) = cos(t) + sin(e) + sin(t) cos(t) Solution: Wehaye c(t) = h(x) +.9¢0)] PLIOA & x= $Isy-s6-0] Pii62 Given :x(0) = cost) + sin(e + sin(9 cox) Pi.t63 We ean obtain x(t) fom x(t) by eplacingt by = cos(-t) +sin(-4)+sin(-t) cos (-1)trtroduetion| a e(-t) = cost) sin) - sin(t) eos(0) PIA6d Substitoting eqn. P1.16.3 & PIL64 ineqn, PI.L6.1 we get, et) = c08(0) Substivuting eqn. P1163 & PL.164 in P1162 we pet, g(t) = sing) [1+e05(0)] Example 1.17 Obtain the even and odd components of the sigeal x(€) = (1-+e') c03! (109) Solution: Wehave a(t) =5lx© + 8-0] PLAT & PLaT2 x(t) = (Lr) cos! (10) PLATS. a(t) = [+G0)'] e08° [10(-4)] att) = (I-#) e051 (109) prita Substituting eqn, P1.17.3 & PLU74 in egn. PLITA we get. s(t) = cost) Substituting eqn. PLI7.3 & PLI74 in eqn. PLI7.2 we get, © cos (104) a wee Example 1.18 Find the even and odd part ofthe signal x(t) = L4e+3e-+ 5-498 Solution; Wehave x =f) +e] PLAS & (9 =Zlst)-xG0] PLA? Given: x(t) = 1404324504984 PLABS. x(t) 14 43-0459 49001 coxa) = Tet 32.5498 piss Substituting eqn, P1.18.3 & PL.18.t im eqn, PLIBAL we get, a(t) = 432498 Substituting eqn, PI.18.3 & PLI84 in eqn. P1182 we get, x(t) = (L450) eae ee Example 1.19 Determine and sketch the even and odd parts of the signal shown in Fig. PL.19 below. Shy +x-9) Solution: Wehave — x( 1 1 Ea) fot) wt38 Signals & Systoms aw (-1)] The gual 4 2(9 1 obeained “The sgn g(t) consists 2 eons fiom aft) by mliphine i strength ( at all ‘Yas shown helow in Fig PLDT bao re ° ‘ 2 Fig. 4.10.41 obuan snilarly. che sagtal 5 s(t) 9 obtained by caking die narror umagy: of (9) aor nnitiplving its serenuth by | at all? as shown helowe in Bip. PI 19.2 [eo 2 7 0 ‘ Fig. P1.192 Adkdiing a(t) ond atch ose get sg(8) as shown mm Pig, PL IDS, ade) 2 a ° 7 2 Fig, 1.19.3 Similarly, we ave x0) =H -60 1 1 2) =X F609 Substracting .a(-t) from 4x(1), we get, (0) as shown below in Fig, PLI9.4iple 120 Determine and sketch the even in Fig, P120. 29 dott Fig, P1194 and odd components of the signal x() shown aw Fig. P1.20 Fig. 91.209 shown below ut Fig P1204 at) Fig. ?1.903 Fig P1204‘Signals & Systems 40 Example 1.21 Deters signal x(n) shown in Fig, PL2L x(n) x(n) + x(-n)] Solution: We have x,(1 (o) +e xn) ‘The signal a,(n) consists 2 terms red from a(n) by multiplying the sary s(n) and $x(en)]. The signal $x(n) is -s of x(n) by $ at all ‘w’ as shown in a(n) pid | a [| 1 Fig. P1214 u fy 4'] a4 * od from x(n) by taking the mirror image of x(n) Similarly the signal 4x(-n) is obtaine cect x(n), then multiplying its sample values by Jas shown in Fig. PL21.2. fxn) TULL Tie eas im 0 a i Fig. P4.91.2‘reroaetion Adding $.x() nnd 4 (em). ¥e goU asa) as showin n Pig, PLL S bg Sate teres fig rit ce ae in se Yenatarly, h ‘lt zt (at) = f-1)] sto) = n= Auten x(n) from $2(n). we get x,(n) ay shown in Fig. P1214 si seal tte Pid Example 122 Dewermine and draw the even and odd parts of the discrete-time signal e(n} shown in Fig, P12, = ag x(n) Foss Geto0tes soos teeter e sa set es Fig P92 Solu | aa $x00) 15, | os 5, - | Fig. 1.00.1Sgaals & Systems fig. P1.09.9 Ackding fatn) and $x(-n). we get at) assown in Fag P1229 aq ln) Tee aseea of es Fig. #1923 Suburacting }{-n) frorn bn). we wetxn) as shown in Fig, PLI24 s(t 1 05, 3 3 re en) sisters os os) \ Fig. P1294 _ OEE Example 12 Find the odd and even part ofthe signal x(@) shown in Fig. P1.23 below x0) Fig. P1.23,Introauction Solution ‘Adding sygnats shown in Fig. P123, Fig P1233 in Fig, PL 23.4 Yr x0) 5-4 4 o a capone Fig, P1.93.1 axe) t ais fig, P1932 1 & Fig, PI.23.2 . we get x(t) and is shown in alt) Fig, 71.933 Subwactng signal shown in Fig, PL 23.2 froma Fig, P123.1, we get s(t) and is drawn Fig. P1934 Example 1.26 Fig, 1.24) and (b) shows part ofthe signal) and its even part e(0) respectively fort 2 Oonily. x(t) and even part x(t) for <0 is not shown, Complete the plots of s(¢) and s(t), Also draw the odd part of x(t) fie. x(t]Sohution : Signals & Systems x(t) (0) 0 1 @ Fig. 71.94 co) We know that even part of any signal is symmetric about t=0, Thus, we have the complete even part x,(0) a8 shown in Fig, P1.24.1 below elt) 2 2 7 ° 4 2 Fig. P1.24.1 We know thar (9 = (0 +20 {c) for €2 0 is obtained in such away that if we adds ¢20 shown in Fig. ?1.24(). Now, a st) and x,() FOEE2 0. we shonld get the given x(0) for a(t) for €20 is shown in Fig, PL242 Fig, P1.24.2Uf wwe add the signal shown in Fup, PI 24(b) and PLL242, we get sft) for 62 0 ts shown in Fig, P1.24(a). Now, we know that odd pare of any real signal is antyymmeime about t=0. Fherefore. the complete x,(t) is shown in Fig. P1.24.3 below Fig. P1.24.3 Now the complere x(t) 1s obtamed by adding complete xe(¢) and (9) as shown in Fig. PI 24.1 and P1243. tis shown in Fig. P1 24.4 below. Fig. P1244 Example 25 Find and sketch the even and ode components ofthe signal x) =P ae) 8 = (0.79) u@) : $80) = F(G.7I we) - Heit) = $ (0.779) u(-t) ; Solution: Given (0) 4& Signals & Systeme “The sketch of Ex(e) and 4x(-9 aie shown an Pig. PI.25.1(a) & (b) below v.28 Yair) fig. P1.95.1 Adding Fig, PL.25.1 (a) and (b) we get s(t) as shown in Fig PL.25.2 below jo @ (e)Iraroduction 4a Suberacting Fig, P1251 (h) from (al, we get_x,(6) a8 shown in Pig. P1.25.3 below. salt) Fig. P1952 I Example £26 Prove that — fatoae = 2 frie Ext) seven m — feowo ee odd Solution : (1) To prove fave HT a(@ seven Luis = fate a , = fos front = “ = fo ae frovee48 Signals & Sysions Prt c= = tise erm, f= = frcoaes fre = fecoae + force | sq seven wo = 000 (i) To prove [xinide = 0 al x(t) as ade LS = fecost u 1 fron + favs 0 = facon + feooee ‘ Pur c= xn) + gfe tn) 1 n=l = x(n). = (0) +) {x(n) + 9¢-m)} we PLDI n=1 For odd signalx(0) ="0 & x(-n) = -x(n). Substituting these in eqn. PI.27.1. we get, = pr) = 0 > {e(n) -x(n)} a=1 =o Hence the proof | Example 1.28 Show hat if x(n) is an odd signal and s(n) is an even signal, then x(n) p(n) isan odd signal. Solution = Consider y(n) = x(n) . x(n) P1281 yen) = xy(-n).x4(-n) ha Baa Given: sey(n) is odd, 1(-n) = -x,(n) P1283 | & y(n) seven. yen) = a(n) ow BABA : Substituting eqn. P1.28.3 & P1.284 in eqn, P1282, we get Yeon) = = (8) a(n) = =y(n) ie, y(-n) {n) ive, y(n) satisfies the condition of odd signal y(n) = xy(n) x,(0) is an odd signal.50 ‘Signals & Systems Example 1.29 Let a(n) be an arbitrary signal with even and odd parts denoted by x(n) and x(n) respectively: Show that. Sew = Faroe San Soluion: Unis = 2") neo Zi > {x09 +60) = {a2} + w26n) + 2edn) -K0)} = Scio | SE ston + BH dey. we PLIOA nee te In eqn, P1291, the last term contains the product of even and odd signal respectively In example 1.28 we proved thatitresultsin an odd signal. By applyingthe equation proved in example 1.27, this term is equal to zero. > “ Savy + Sato) Hence the proof, 4.6.3 Periodic and Non-periodic signals ‘A continuous-time signal x(:) is said to be periodic iP satisfies the condition, x(t) = (047) Hforall't 54) where ‘T" isa positive constant IF the condition in equ. 1.54 is satisfied for T=T.. then itis also satisfied for any ‘Tent, where n=1,2,3, ‘The smallest value of “T’ that satisfies eq. 1.54 is called the famdarental period of (0. This fundamental period is the time taken by the signal x(9) to complete its one ‘cle. The reciprocal of the fincamental period ‘T” is known as the fundamental frequency of the signal“Invedvetion LOL BHPSE ‘The fundamental angular frequency ‘e" is given by FE Gants) o= mas ‘Any continuous-time signal x(t) which does not satisfy eqn ‘non-periotic or aperidic sign. Examples for periodic and non-periodse continnovs-tin and 1.31 respectively, e signal is shown in Fig, 1.30. +) 4a 3 2 7 of tf @ 3 4 Fig. 1.30 A continuous tine peo signel with the fundamental period T=2 x@ Fig. 1.31 continoustime sonpereic signal Similarly a discrete-time signal x(n) i sac to be peed iFit satisfies the condition, x(n) =x(atN) forall a? ( where N is a positive integer, The smallest value of IN which satisties eqn, 1.55 38 called the funudarnentt period of the signal x(n). ‘The fundamental angular feequeney of =) is given by, 2n = (radians) Any dlscrete-time signal x(n) which doesnot satisfy eqn. 1.55 is callee non-periade or aperiodic seal, Examples for periadie and non-periodie discrete-time signal is shown in Fig. 132 and Fig, 1.33 respectively62 ‘Signals & Systems Fig. 1.32 Adiscretetime posidic signe with fundamental period N= 6 a(n) 4 05, 9.25 Pa setores +e oF - Fig. 1.33 Adserete-tmo non porode signal Exams eS Feample {.30 Determine whether the continuous time signal x(t) = [cos(2at)?is periodic oo not. Ifperiodic, find the Fundamental period T. Solusion: Given: x(1) = [cosa]? 1 = HLL +cos re) 1 1 Sa peos ne In the above expression for x(t), the first term is constant (ie. average valve). The second term is a cosine signal with maximum amplitode of 1. Comparing vith cos @,t, we im x have @, “The signal cost is periodic with period T = 2% =20 2 on PIBOL P1302 We have ‘Comparing P130.1 and P1.30.2 we have, Fundamental period T=05 sec: xi) Verification => x(t+T)