Sig & Sysons oe Se mayo YH FS. nA 5 is . riots |e 1 x(t)re "edt | C 23.1 ae ole f ; 3.1) © The term inside the square bracket in Equation (2.3.13) is nothing but “C,’ Pe Y GG But C,+C* = 1¢,? Pe Ic,P ... Proved oes (2.3.14) e " 7 i il average power of the signal © The interpretation of the Parseval's theorem is that the total » power of the signal x(t) can bo found by squaring and adding the heights |C,,| of the amplitude lines in the spectrum of the periodic signal x(t). ‘* Thus the Parseval’s theorem implies “superposition” of the average powers. 2.4 Fourier Transform : * Till now we have seen how to represent the periodic signals extended over the interval (co, «), using the fourier series. Non-periodic time limited signal can also be represented by the fourier series. * However the non-periodic signals which extend from —o to oo can be represented. more conveniently using the “Fourier Transform” in the frequency domain. (b) Continuous spectrum as f, +0 Hines at fy 5 2f sme Fig. 2.4.1 11 is possible to find the fourier transform of periodic signal as well, For the periodic signals T, ><. Hence the frequency f, T, 7.0 Therefore the difference between the spectralystems (PTU) signals & SYS (PTU) 2-35 Signal Analysis using F. S. & F. T. 24d What does a Fourier Transform do ? ‘Any signal is build up by addition of clementary si i i i have different amplitudes and relative phases, "> S&T Which are at different feqvencies, Using the fourier transform we can plot the amplitude and phase spectrums whi i t : ; a ich provide us atte information about amps and relative pases of sc clemenlary sigaais, Thus fourier transform can be used for the “anal fa signal. It i transformation from the time domain to frequency demain,» | Sana Mt IS used for ition of Fourier Transform : 24.2 Defi rhe fourier transform of a signal x(t) is defined as follows : Fourier tansform: — X(o) = fxctye at a w(2.4.1) ORF XC) S [x (ewer ‘These equations are known as “analysis” equations. 2.4.3 Definition of Inverse Fourier Transform : The signal x(t) can be obtained back from fourier transform X(f) by using the inverse fourier transform. The inverse fourier transform (IFT) is defined as follows : Inverse fourier transform : x(t) =2 J x(o) edo st 2 ij (24.2) OR x(t) = [ x(t) ear Representation : ‘The signal x(t) and its fourier transform X (£) form a fourier transform pair which can be Tepresented as, x(t) & xc) G43) The other way to represent is as follows : . X(f) = F[x(t)] (244) or it can be represented as, x(t) = FU'EX(£)] The fourier transform is a complex function of frequency f. Therefore itin the complex exponential form as follows : (2.4.5) it is possible to express ety = 1x(H1ed2® (2.4.6) =Signals & Systems (PTU) 2:36 Signal Analysis using F. S. & F. T. In this expression : Ix(f)l O(f) = The phase spectrum. ‘The amplitude spectrum is a graph of amplitude versus frequency. Whereas the phase spectrum is a graph of phase angle versus frequency. 2 nditions for the Existence of Fourier Transform : >>> university Exam - Dec. 2006 Ht} * We have studied the Dirichlet conditions in section 2.2.3. Similarly these conditions should ‘be satisfied by a signal x(t), then only itis possible to obtain the fourier transform of x(t). For the periodic signals the integration is obtained over one period however for the periodic signals, it will be obtained over a range —2o to «0, ‘The signal x(t) will have to satisfy the following conditions so that it’s fourier transform can be obtained : @ The function x (t) should be single valued in any finite ime interval T. Gi) It should have a finite number of discontinuities in any finite interval T. Gil) The function x(t) should have a finite number of maxima and minima in any finite interval of time T. “@v) The function x(t) should be an absolutely integrable function. ‘The amplitude spectrum of x(t) and That means [1x(t)I dt < © ‘The conditions stated above are sufficient conditions, but they are not the necessary conditions. O47 2.4.5 Amplitude and Phase Spectrums : * The amplitude and phase spectrums are continuous rather than being discrete in nature. Out of them, the amplitude spectrum of a real valued function x(t) exhibits an even symmetry. X(f) = XGf) 248) ©. And the phase spectrum has an odd symmetry. That means, O(f) = -0(-f) (2.4.9) 26// Properties of Fourier Transform : Some of the important properties of the fourier transform arc listed as follows : © Linearity or superposition ‘Area under X (') © Time scaling * Differentiation in time domain © Duality or symmetry © Integration in time domain © Time shifting ©, Conjugate function © Frequency shifting * Multiplication in time domain ° (Multiplication theorem) © Area under x(t) © Convolution theorem, Let us understand these properties one-by-one,signals & Systems (PTU) 2.37 Signal Analysis using F. S. & F. T. 25.1 Property 1 : Linearity or Superposition : F a Ex, (t) > X, (£) and x, () <5 X,(£) represent the fourier transform pairs then for all constants such as a, and a, we can write, [" x (+a, x, (0) | ar 1X (1) aX, (1) ] (2.5.1) «That means the linear combination of inputs gets transformed into linear combination of their fourier transforms. + This property can be used to obtain the fourier transform of a complicated function by decomposing it in the form of sum of elementary functions. Proof of the Property : | ¥ (0 #8,%(1)] =f [ax rar (oem a fa, x (tye Pat #fayacn-2 ae = a,X,(f) +a,X,(£) 25.2 Property 2 : Time Scaling : Let x(t) and X(f) form a fourier transform pair and I . Then the time scaling property states that, Fo x(ot) <> 4X (fla) (2.5.2) Meaning : (1) x (at) represents a time scaled signal and X (f/ c) represents the frequency scaled signal or scaled frequency spectrum. Q) For a <1, x(at) will be a compressed signal but X (f/ 0.) will be an expanded version of X(f). : @) And for cw > 1, x(at) will be an expanded signal in the time domain. But its fourier transform X ({/ 0.) represents a compressed version of X (). ‘Thus compression in-the time domain results in expansion of frequency spectrum whereas lexpansion in the time-domain results in compression of the frequency spectrum. Proof of scaling property : “ce” being a constant, can be-positive or negative. ic. & > 0 or o < 0. Let us find the F.T. Considering both the possibilities.Signals & Systems (PTU) 2:38 is using F.S. & F. T. @ “a” positive (a>0) F[x(at)} = fxatye a Substiute t = at, thon dt = adt dt = ae @ F[x(at)] = fx(z) Pa ot = Lh fare Pg, = x(a) 2.53) a Gi) a inary 3) Ex. 2.5.2 : Find the fourier transform of the exponential pulse shown in Fig. P. 2.5.2(a). Also find the amplitude and phase ‘spectrums for the same. x(t) 1 0.3678 st Va t Fig. P. 2.5.2(a) Soln. : The pulse shown in Fig. P. 2.5.2(a) can be represented as, x(t) = e™ fort co =0 for t>0 nl) It can also be represented as, x(t) =e u(-t) +) It's fourier transform is given as, . i ol : ° Xf) = fe u(-ty-e Ph at = feet oP Gy = Flam statyeSignals & Systems (PTU) 2-41 Signal Analysis using F. S. & F. T. ~ Ci eee 1 ou 1 1 xf) = —1_/.?. oe = . Tanpanl* =] Canjanty 1-9) = Ca rjanty ~@) sis is the desired result. Thus, eMuct & A 4) (a= j2nfy To find the amplitude spectrum : ‘The amplitude spectrum is denoted by |X (f)| and is given by, V2. IX(f)1 = [(Reat partof X(f))*+(Imaginary part of x(t)? ] 1 1 a+j2nt) a+ j2nf We have found out that X(f) = ———=—__ Co+i2nf) __arjont— ene ) = Canitaty Cain)” (at iomt) ~ a +(2nf yp Real part of X(f) = =“ > KC) pen of XC) = Fave 2 Ae nf .d imaginary part of X(f) = = > O) and imaginary pé (f) eae 7 Therefore the amplitude spectrum is given as, 2 Ix(f)l = | —Sa5 Oo [ave IX(f)l = 1 AT) (era Py? Substitute different values of “f” in Equation (2.5.6) and plot the amplitude spectrum. It looks as shown in Fig. P. 2.5.2(b). To obtain the phase spectrum : ‘The phase spectrum is denoted by 0 (f) and is expressed as, oct) = tan? [sare part of aml Real part of X(f) a 6(f) Radians | bet] (c) Phase spectrum of e“u(-t) (b) Amplitude spectrum of eu (—t) Fig. P. 25.2Signals & Systems (PTU) 2-42 Signal Analysis using F. S. & F, Substituting the values of real and eH values of X(f) we get, Ba af The phase spectrum: OCF) = tan! Substitute different values of “I in the Scie (7) and plot the phase spectrum. It looks shown in Fig. P. 2.5.2. Ex. 2.5.3: Obtain the fourier transform of the double exponential pulse shown in the Fig. P. 2.5.3(a). Fig. P. 2.5.3(a) : Double exponential pulse Soln. : ‘The signal x(t) is the sum of two exponential pulses ie. X(t) = x(t) +x, (t) = eMu(-ty +e" uty Taking the fourier transform of both sides, F[x(t)] = Fle“u(—t)]+Fle~“ uty] The fourier transforms of the two terms on the RHS have already been found in the Ex. 2! and Ex. 2.5.2. Substituting them we get, X(f) o-j2nf a +jont = SA janf +o —jant a + (Qnf )? 20 X(f) = —24 __ o? + (nt > @ The amplitude spectrum 1X(f)| = ——2%__ o + (Qnt Gi) The phase spectrum e(f) = 0. The amplitude and phase“spectrums are plotted as shown in Fig. P, 2.5, 3(b). 0 0| signals & Systems (PTU) 2-43 Signal Analysis using F. S. & F. T. (b) Amplitude spectrum Amplitude spectrum of) (© Phase spectrum Phase spectrum ' 0 Fig. P. 2.5.3 Obtain the fourier transform of the delta function shown in Fig. P. 2.5.4(a). Ex. ea x(t) = 3(t) o t Fig. P. 2.5.4(a) : Delta function Soln. : @ By the definition of fourier transform, x(f) = Jxctye Pat = f5ctye PM at wl) We cannot substitute the value of 6(t) directly in the Equation (1) because it is infinitely 0. Therefore let us use the sifting property of the delta function: large at (i) Sifting property of delta function : The sifting property states that frceys(t-tyyat = FC) 2) Let us use this property in Equation (2.5.1) as follows : (ii) In Equation (2.5.18) assume that ty = 0 and f(t) = Caos xf) = feP**.5(1-0) +. by using Equation (2). xc) = eg butt, =0 xf) = PML QB)Signals & Systems (PTU) 2-44 Signal Analysis using F. S. & F, 7 Thus 8(t) <1 The amplitude spectrum of the delta function is as shown in the Fig. P. 2.5.4(b) This shows that the delta function contains all the frequencies from ~oo to oo with equal amplitudes. The fourier transform of a delta function is a de signal. 7 ; Fig. P. 2.5.4(b) : Amplitude spectrum ; delta function Ex. 2.5.5: Obtain the F.T. of the antisymmetrical pulse shown in the Fig. P. 2.5.5, Fig. P. 2.5.5 Soin. : The antisymmetric pulse can be represented as, x(t) = 6% t>0 = lll t=0 = -e™ t<0 Therefore the fourier transform is given by, o oe ° tia “i i F[x(t)] = f ye" Lm s fete a . =o o” o* = fel -Part ge fer Coteant a ot ale 1 eee = e + aaa « ina = Teepe L? + go-to [e"*-e"]i = ‘Signals & Systems (PTU) 2-45 Signal Analysis using FSF. T. rly sa aja = jon (a-j2nf) ” (a +j2nf) a? +(2nf)* | 1 X(f) =, aia 225.1) a + (2nf)? oe Ex. 2.5.6: Fihd fhe fourier transform of the signum function shown in Fig. P. 2.5.6. (May 2005, 2 Marks) x(t) = sgn (t) 1 Fig. P. 2.5.6 : Signum function Soln.: In the Ex. 2.5.5, we have found FT of an antisymmetrical exponential pulse. The antisymmetrical pulse can be represented as, x(t) =-e7'*!*sgn (ty Al) ‘Therefore as proved in Ex. 2.5.5, the FT of the antisymmetric exponential pulse is given by, slalt —j4nf Fle!" sent) ] = SG 2 [ ea ] oe +(2nf)? 2 In the Equation (2), substitute a = 0 to get, ~jdnf Fle sgn(t)] = —GS [e® sgn(t) are 2 el F[sgn(t)] at wum function sgn (0) <> : wAns. Ex. 25.72 Obtain the fourier transform of a unit step function. (Dec. 2005, 5 Marks) | - Soln, : A unit step function is mathematically defined as, a u(t) = 1 for t>0 wan 0 elsewhereSignals & Systems (PTU) 2-46 Signal Analysis using F. S. & F.T, Using the definition of the fourier transform we get, Fray} = Juco" But unit step function is present only for t 2 0 To mient 1p zat” Flu(t)] = ine Feat a Tale | op sei 0 eS Saal * -e'] 1 =. Fa lO 1j= jint oe Ans, u(t) <> jet Ex. 2.5.8 : Obtain the fourier transform of a rectangular pulse of duration T and amplitude A as ‘shown in Fig. P. 2.5.8(a). xit) A =e lo T2 ‘ Fig. P. 2.5.8(a) : Rectangular pulse Soin. : ‘The rectangular pulse shown in Fig. P. 2.5.8(a) can be expressed mathematically as, rect (UT) = A for -TA ATsinc ({T) ‘Thus the rectangular pulse transforms into a sinc function. " sinc (fT) Amplitude spectrum : The amplitude spectrum of the rectangular function is as shown in Fig. P. 2.5.80). xo] ‘Amplitude spectrum Fig. P. 2.5.8(b) : Amplitude spectrum of a rectangular pulse <. AT sinc Q) = AT As we already know, sinc (0) = 1 The sine function will have zero value for the following values of “fT”: sinc (1) = 0 for fT = 41, £2, £3, nm The phase spectrum has not been shown as it has zero value for all the values of £. on) Q@ (3)Signals & systems (PTU) 2-48 Slonal Analysis using F. 8. & & To absorb negative values of IX(f)Lin the phase shift : ‘The negative amplitude of the amplitude spectrum 1x introducing a phase shift of negative phase shi for posi ‘introduced in order to maint (EI can be made positive by S as shown in Fig. P. 2.5.8(). 4 hift for the negative frequency is £180" in the phase spectrum, This i itive frequency and positive phase sl tain symmetry of the phase spectrum. x] Phase spectrum Phase shift of -180° to absorb the negative Values of |X(f)| and to make |X()| positive Fig. P. 2.5.8(c) : Amplitude and phase spectrums for a rectangular pulse. Negative values of Ixce )I have been absorbed in the additional phase shift of + 180° in the phase spectrum [note i The choice of the phase shift of + 180" is completely arbitrary. 25.3 Property 3 ; Duality or Symmetry Property : This property states that, t xO Ax Lr then t and f can be interchanged, Meaning : (1) In the term x(t), “x” represents the Shape of the signal and “t” shows that the variable is time.