9.4 Partial fractions 517
11 Partial fraction expansions have their principal applications in electrical
engineering and elsewhere where complex numbers (numbers of the form a + ib,
where a and b are real and i is the imaginary unit for which i2 = -1) are system-
atically used. When complex numbers are used, it is not necessary to bother
with factors (like x2 + 1 and x2 + x + 1) that cannot be factored into real
linear factors. Theory and applications can be based upon a version of the
fundamental theorem of algebra which says that if
(1) P(x) = aoxn + alxn-1 + ... +an-1x + an,
where n is a positive integer and ao ; 0, then there exist k different numbers
x3, x2,. , xk (which are not necessarily real) and k exponents p1, p2i
pk (which are positive integers) such that
(2) p1+p2+ +pk=n
and
(3) P(x) = ao(x - x1)"(x - x2)P2 ... (x - xk)Pk.
This theorem is a simple corollary of general theorems that appear in the theory
of functions of a complex variable. Less sophisticated but more complicated
proofs can be given. The numbers x1, x2, , xk are said to be zerosof the
polynomial of multiplicities p1, p2,- ' -, P. It can be proved that if Q is a
polynomial of degree less than that of P, then Q(x)/P(x) is representable in the
form
( Q(x)_ Al., 41,2 41,3 41,P1
) P(x) (x - xl) + (x - x1)2 + (x - x1)3+ +(x- xl)PI
42,1 42.2 42,3 42,Pi
(x - x2) (x - x2)2 + (x - x2)3+
+(x - x2)P!
+
4,1
+
4.2
+
4,3
+ +
4k.k
(x - xk) (x - xk)2 (x - xk)3 (x- ,xk)Pk
where the pi constants 141,1, 41.2,', - ,41,P, "go with the powers of the factor
(x - xl)," the P2 constants 42,1, 42,2, ... , 42.P, "go with the powers of the
factor (x - x2)," etcetera. Supply and demand do not generate heavy traffic
in proofs of this partial fraction theorem, but we can pause to look at the three
potent identities
(5)
(6)
and
(7)
x R(x) (x - a) + a R(x)
x - a S(x) x - a S(x)
1 R(x)^1 (x - a) - (x - b) R(x)
(x - a) (x - b) S(x) b - a (x - a)(x - b) S(x)
x R(x)^1 b(x - a) - a(x - b) R(x)
(x-a)(x-b)S(x) -h-a (x-a)(x-b) S(x)