Lecture 2: Signals and Systems, Part I

2 Signals and Systems: Part I In this lecture, we consider a number of basic signals that will be important building blocks later in the course. Specifically, we discuss both continuoustime and discrete-time sinusoidal signals as well as real and complex exponentials. Sinusoidal signals for both continuous time and discrete time will become important building blocks for more general signals, and the representation using sinusoidal signals will lead to a very powerful set of ideas for representing signals and for analyzing an important class of systems. We consider a number of distinctions between continuous-time and discrete-time sinusoidal signals. For example, continuous-time sinusoids are always periodic. Furthermore, a time shift corresponds to a phase change and vice versa. Finally, if we consider the family of continuous-time sinusoids of the form A cos wot for different values of wo, the corresponding signals are distinct. The situation is considerably different for discrete-time sinusoids. Not all discrete-time sinusoids are periodic. Furthermore, while a time shift can be related to a change in phase, changing the phase cannot necessarily be associated with a simple time shift for discrete-time sinusoids. Finally, as the parameter flo is varied in the discrete-time sinusoidal sequence Acos(flon + 4), two sequences for which the frequency flo differs by an integer multiple of 27r are in fact indistinguishable. Another important class of signals is exponential signals. In continuous time, real exponentials are typically expressed in the form cet, whereas in discrete time they are typically expressed in the form ca". A third important class of signals discussed in this lecture is continuoustime and discrete-time complex exponentials. In both cases the complex exponential can be expressed through Euler's relation in the form of a real and an imaginary part, both of which are sinusoidal with a phase difference of 'N/2 and with an envelope that is a real exponential. When the magnitude of the complex exponential is a constant, then the real and imaginary parts neither grow nor decay with time; in other words, they are purely sinusoidal. In this case for continuous time, the complex exponential is periodic. For discrete Signals and Systems 2-2 time the complex exponential may or may not be periodic depending on whether the sinusoidal real and imaginary components are periodic. In addition to the basic signals discussed in this lecture, a number of additional signals play an important role as building blocks. These are introduced in Lecture 3. Suggested Reading Section 2.2, Transformations of the Independent Variable, pages 12-16 Section 2.3.1, Continuous-Time Complex Exponential and Sinusoidal Signals, pages 17-22 Section 2.4.2, Discrete-Time Complex Exponential and Sinusoidal Signals, pages 27-31 Section 2.4.3, Periodicity Properties of Discrete-Time Complex Exponentials, pages 31-35 Signals and Systems: Part I 2-3 IDAL SIGNAL TRANSPARENCY 2.1 Continuous-time sinusoidal signal indicating the definition of amplitude, frequency, and phase. 27 0W0 t iallest To TRANSPARENCY 2.2 Relationship between a time shift and a change in phase for a continuous-time sinusoidal signal. Signals and Systems ----------- x(t) = A cos wo t TRANSPARENCY 2.3 Illustration of the signal A cos wot as an even signal. 2r A 0 A WO r V Periodic: x(t) =x(-t) Even: A cos (w t - -ff) - x(t) = A sin wot A cos [ w(tTRANSPARENCY 2.4 Illustration of the signal A sin wot as an odd signal. Periodic: x(t) = x(t + TO) Odd: x(t) =-x(-t) )] Signals and Systems: Part I TRANSPARENCY 2.5 Illustration of discrete-time sinusoidal signals. Time Shift => Phase Change A cos [920(n + no)] = A cos [Mnn + E2 0 n0 ] TRANSPARENCY 2.6 Relationship between a time shift and a phase change for discrete-time sinusoidal signals. In discrete time, a time shift always implies a phase change. Signals and Systems 2-6 p=0 x[n] =A cos 0 n TRANSPARENCY 2.7 The sequence A cos flon illustrating the symmetry of an even sequence. n 71' 90 8 0 8 0ee .00 even: x[n] = x[-n] A cos (En x[n] = TRANSPARENCY 2.8 The sequence A sin flon illustrating the antisymmetric property of an odd sequence. A sin ) E2n A cos [W2(n - no)] n 0 =I 000 odd: x[n] = -x [-n] Signals and Systems: Part I 2-7 Time Shift => Phase Change A cos [20(n + no)] = A cos [Mon + 9 0 n0 ] <= Time Shift A cos [92 0 (n + no)] = TRANSPARENCY 2.9 For a discrete-time sinusoidal sequence a time shift always implies a change in phase, but a change in phase might not imply a time shift. Phase Change A cos [2 n ++J x[n] = A cos (Wn+ #) Periodic? x[n] = x [n + N] smallest integer N = period A cos [20(n + N) + #] = A cos [20 n + 2 N + #] integer multiple of 27r ? Periodic = > 920 N 27rm 27rm 0 N,m must be integers smallest N (if any) = period TRANSPARENCY 2.10 The requirement onne for a discrete-time sinusoidal signal to be periodic. Signals and Systems 12 060 I TI TI TI I 'T Go* - a " 1 01 TRANSPARENCY 2.11 Several sinusoidal sequences illustrating the issue of periodicity. In I 1 01 n H,.TT~ ~TI 000 S31 0 TRANSPARENCY 2.12 Some important distinctions between continuous-time and discrete-time sinusoidal signals. 0 6 p= 0 1IT,. 00T0I A cos(oot + #) Distinct signals for distinct values of wo Periodic for any choice of o ese A cos(E 0 n + G) Identical signals for values of Eo separated by 27r Periodic only if o _ 21rm N for some integers N > 0 and m Signals and Systems: Part I SINUSOIDAL SIGNALS AT DISTINCT FREQUENCIES: Continuous time: x 1 (t) = A cos(w 1 t +) x 2 (t) If = A cos (w 2t +) w2 1 Then x 2 (t) # x1 (t) Discrete time: x, [n] = A cos[M, n + ] x 2 [n] = A cos[2 2 n + #] If 22 TRANSPARENCY 2.13 Continuous-time sinusoidal signals are distinct at distinct frequencies. Discretetime sinusoidal signals are distinct only over a frequency range of 2,. 1 + 27rm Then x 2 [n] = x, [n] REAL EXPONENTIAL: CONTINUOUS-TIME x(t) = Ceat C and a are real numbers X(t) TRANSPARENCY 2.14 a >0 C a <0 Time Shift <=> Scale Change Cea(t + to) = Ceato eat Illustration of continuous-time real exponential signals. Signals and Systems 2-10 REAL EXPONENTIAL: DISCRETE-TIME TRANSPARENCY 2.15 Illustration of discrete-time real exponential sequences. x[n] = Ceon = Can C,a are real numbers L a>0 a< n a l a>0 n Jal <1 COMPLEX EXPONENTIAL: CONTINUOUS-TIME TRANSPARENCY 2.16 Continuous-time complex exponential signals and their relationship to sinusoidal signals. x(t) = Ceat C and a are complex numbers C= ICI ej 6 a= r + jo x(t) = 0 ejo e (r + jcoo)t = ICI ert ej(wot + 0) Euler's Relation: cos( 0t + 0) + j sin(wot + 0) = ej(wot + 0) x(t) = ICi ert cos(cot + 0) + jl|C ert sin(wot+ 0) Signals and Systems: Part I 2-11 TRANSPARENCY 2.17 Sinusoidal signals with exponentially growing and exponentially decaying envelopes. COMPLEX EXPONENTIAL: DISCRETE-TIME x[n] = Can C and a are complex numbers C = ICI eji a= 1al ei x [n] = C ej (lal ejo) =IC 1al n e j( n on + 0) Euler's Relation: cos(2 0 n + 0) + j sin(&20 n + 0) x[n] = ICI lal n cos(92on + 0)+ j ICI|al n sin(92on + 0) |al = 1 => sinusoidal real and imaginary parts Ce jon periodic ? TRANSPARENCY 2.18 Discrete-time complex exponential signals and their relationship to sinusoidal signals. Signals and Systems 2-12 |al > I TRANSPARENCY 2.19 Sinusoidal sequences with geometrically growing and geometrically decaying envelopes. lal