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

# O therwise t he p rocedure remains t he s ame t he p

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: e B .ll Expand F (x) into partial fractions i f which agrees w ith t he p revious result. F (x) = 3 x 2 + 9x x2 A Mixture o f t he Heaviside "Cover-Up" and Short Cuts Here m = n = 2 with bn = b2 = 3. +X - 20 6 = 3 x 2 + 9x - 20 (x - 2)(x + 3) Therefore, I n t he above example, after determining t he coefficients ao = 2 a nd k = 1 by t he Heaviside m ethod a s before, we have 4 x 3 + 1 6x 2 + 23x + 13 2 al a2 1 - -;---:-.,-;;,-;----,-- = - - - + - - - + - - + - (x+l)3(x+2) ( x+l)3 ( x+l)2 x +l x +2 T here a re only t wo u nknown coefficients, a l a nd a2. I f we m ultiply b oth sides of t he above e quation b y x a nd t hen let x - > 0 0, we c an eliminate a 1 . T his yields in which kl = 3x 2 + 9x - I 20 = 12 + 18 (2 z =2 and 2 k2 = 3x + 9 x - 20 ~ I 20 = + 3) 5 = 27 - z =-3 ~ =2 27 - 20 ( -3-2) = - 20 = 4 -5 Therefore, F (x) Therefore, ( x - 2)(x 4 x 3 + 1 6x 2 + 2 3x + 13 2 al 3 1 - -,.--.....,...,,-;----- = - - - + - - - + - - + - (x + 1 )3(x + 2) ( x+l)3 ( x+l)2 x +l x +2 T here is now o nly one unknown a I, which c an b e r eadily found by s etting x e qual t o a ny convenient value, say x = O. T his yields ¥ = 2 + al + 3 + ~ =} Improper F (x) with m 8 .5-6 = 3 + _ 2_ + _ 4_ x- 2 x +3 • Modified Partial Fractions F () = x 5x2 + 2 0x + 18 ( x + 2 )(x + 3)2 --...,.-,.----,-= Dividing b oth sides by x yields =n A g eneral m ethod of handling an i mproper function is indicated in t he beginning of t his s ection. However, for a special case where the n umerator a nd d enominator p olynomials of F (x) a re of t he s ame degree (m = n ), t he p rocedure is t he s ame as t hat for a p roper function. We c an show t hat for 20 + 3) O ften we r equire p artial f ractions of t he form (x~f.)r r ather t han ( x_\.jr. T his c an b e achieved by e xpanding F (x)/x i nto p artial fractions. Consider, for example, al = 1 which agrees w ith o ur e arlier answer. B .5-5 = 3 x 2 + 9x - F (x) 5x 2 --= x x (x + 2 0x + 18 + 2 )(x + 3)2 E xpansion of t he r ight-hand side into p artial f ractions as u sual yields F (x) = 5 x + 2 0x + 18 = a l + ~ + _ a_3_ + _ a_4_ x x (x + 2 )(x + 3)2 X X+2 ( x + 3) ( x + 3)2 Using t he p rocedure discussed earlier, we find a l = 1, a2 = 1, a3 = - 2, a nd a4 = 1. T herefore, 2 F (x) = .!.+_1_ _ _ _+_1_ 2 x x x+2 x+3 ( x + 3)2 Now multiplying b oth s ides by x yields t he coefficients k l' k 2,"" k n a re c omputed a s if F (x) were proper. Thus, x 2x x F (x) = 1 + - -2 - --3 + - ( 3)2 x+ x+ x+ T his expresses F (x) as t he s um o f p artial f ractions having t he form (x~f.)r. For q uadratic o r r epeated factors, t he a ppropriate p rocedures discussed in Secs. B.5-2 or B.5-3 s hould b e u sed as if F (x) were proper. I n o ther words, when m = n , 34 Background B .6 Vectors and Matrices An entity specified by n numbers in a certain order (ordered n-tuple) is a n n-dimensional v ector. Thus, an ordered n-tuple ( Xl, X 2, . .. , X n) represents an n-dimensional vector x. Vectors may be represented as a row ( row v ector): X =[XI X2 X nl o r as...
View Full Document

## This note was uploaded on 04/14/2013 for the course ENG 350 taught by Professor Bayliss during the Spring '13 term at Northwestern.

Ask a homework question - tutors are online