Signal Processing and Linear Systems-B.P.Lathi copy

# In this approach the impulse function is defined by

This preview shows page 1. Sign up to view the full content.

This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: generalized function definition of a n impulse. Because t he u nit s tep function u (t) is discontinuous a t t = 0, i ts derivative d u/dt does not exist a t t = 0 in the ordinary sense. We now show t hat 72 1 I ntroduction t o Signals a nd Systems 73 1.4 Some Useful Signal Models t his derivative does e xist in the generalized sense, a nd i t is, in fact, o(t). AB a proof, l et us evaluate t he integral of (du/dt)&lt;{I(t), using integration by parts: oo du &lt;{I(t) dt = u(t)&lt;{I(t) dt J - 00 00 - 00 - °-1 = &lt;{I(oo) - 1 (a) 00 1 -00 u(t)¢(t) dt (1.25) 00 ¢(t) dt = &lt;{I(oo) - &lt;{I(t)l~ = &lt;{I(O) (1.26) T his r esult shows t hat d u/dt satisfies t he s ampling property of o(t). Therefore i t is a n impulse o(t) i n t he generalized s ense-that is, ~~ = (1.27) o(t) F ig. 1 .21 Sinusoids of complex frequency Consequently 3. loo o(r)dr = u(t) (1.28) T hese r esults c an also b e o btained graphically from Fig. 1.19b. We observe t hat t he a rea f rom - 00 t o t u nder t he l imiting form of o(t) in Fig. 1.19b is zero if t &lt; 0 a nd u nity i f t 2: O. Consequently 1 t T he Exponential Function (7 + j w. e st One of the most important functions in t he a rea of signals and systems is the exponential signal e st, where s is complex in general, given by s= (7 + jw Therefore (1.30a) O (r)dr={O If s· = 1 - 00 = u(t) (7 - jw ( the conjugate of s), t hen (1.29) (1.30b) Derivatives o f impulse function can also b e defined as generalized functions (see Prob. 1.4-10). a nd ;:,. Comparison of this equation with Euler's formula shows t hat e st is a generalization of the function eiwt , where t he frequency variable j w is generalized t o a complex variable s = (7 + j w. For this reason we designate t he variable s as the c omplex f requency. From Eqs. (1.30) i t follows t hat t he function e st encompasses a large class of functions. T he following functions are special cases of e st: E xercise E 1.9 Show t hat + 3)6(t) = ( a) (t3 ( c) e - 2t 6(t) = 6(t) Hint: Use Eqs. (1.23). ;:,. 3 6(t) ( b) ( d) [sin ( t 2 - ~)16(t) = - 6(t) w2 +1 1 w2 +9 6 (w-l) = 5 '6(w-l) \l 1 2 3 4 E xercise E 1.l0 Show t hat ( a) ( c) 1: 1: 6 (t)e- j &quot;,t dt = 1 e - 2(x-t)6(2 - t) d t = e - 2(x-2) Hint: In p art c r ecall t hat 6 (x) is located a t x is at t = 2. \l (1.30c) = O. T herefore 6(2 - t) is located a t 2 - t = 0; t hat A c onstant k = k e Dt (s = 0) A monotonic exponential eut (w = 0, s = ( 7) A sinusoid cos wt (7 = 0 ,8 = ± jw) An exponentially varying sinusoid eut cos wt (8 = (7 ± j w) These functions are illustrated in Fig. 1.21. The complex frequency 8 can be conveniently represented on a c omplex f requency p lane (8 plane) as depicted in Fig. 1.22. T he horizontal axis is the real axis (7 axis), and the vertical axis is t he imaginary axis (jw axis). The absolute value of t he imaginary p art of s is Iwl ( the radian frequency), which indicates t he frequency 74 1 I ntroduction t o Signals a nd Systems 75 1.5 Even a nd O dd F unctions fo ( t) Left h alf plane t- 1 -----1----- Fig. 1.23 An even and an odd function of t. &gt;. ~ I 1....
View Full Document

Ask a homework question - tutors are online