2.

9 Differential and Difference Equation Representatons of LTI Systems 141

Answers:
(a) s[n] = (2 - (1/2)")u(n]
(b) st) e'u(-t) +t (2 - e)u(t)
(c) s(t) = u(1) - u(t - 1)

| 2.9 Differential and Difference

Equation Representations of LTI Systems
Linear constant-coefficient difference and differential equations provide another represen
tation for the input-output characteristics of LTI systems. Difference equations are used to
represent discrete-time systems, while differential equations represent continuous-time sys
tems. The general form of a linear constant-coefficient diferential equation is
N M
dye*4), (2.35)
k=0

where the 4,and the b, are constant coefficients of the system, x(t) is the input applied to
the system, and y(t) is the resulting output. A linear constant-coefficient difference equa
tion has a similar form, with the derivatives replaced by delayed values of the input x(n)
and output y[n]:
N M

Z4ss[n - k] = S b,x(n - k].

k=)
(2.36)

The order of the differential or difference equation is (N, M), representing the number of en
ergy storage devices in the system. Often, N M, and the order is described using only N.
As an example of a differential equation that describes the behavior of a physical sys
tem, consider the RLC circuit depicted in Fig, 2.26. Suppose the input is the voltage source
x(t) and the output is the current around the loop, y(t). Then summing the voltage drops
around the loop gives
d
Ry(t) + a)¿1)d =x(4).
Differentiating both sides of this equation with respect to t results in
1 d d
o) +R() +L) ).

x(t)

F1GURE 2.26 Example of an RLC circuit described by a differential equation.

142 CHAPIER 2 TIME-DoMAIN REPRESENTATIONS oF LINEAR TIME-INVARIANT SYSTEMS

This differentialequation describes the relationship between the current y(?) and the vol.
age x(t)in the circuit. In this example, the order is N = 2, and we note tha the circuit con
tains two energy storage devices: a capacitor and an inductor.
Mechanical systems also may be described in terms of differential equations tha
make use of Newton's laws. Recall that the behavior of the MEMS accelerometer modeled
in Section 1.10was given by the differential equation
O,d
oy(t) + Q d )+ driV(1) = x(1),
where y(t) is the position of the proof mass and x(t) is the external açceleration. This
system contains two energy storage mechanisms-a spring and a mass-and the order
1S again N= 2.
An example of a second-order difference equation is
1
yln] + yln - 1] + Yn - 2] = x[n] + 2x[n - 1], (2.37)

which may represent the relationship between the input and output signals of a system
that processes data in a computer. Here, the order is N = 2, because the difference equa
tion involves y[n - 2], implying a maximum memory of 2 in the system output. Memory
in a discrete-time system is analogous to energy storage in a continuous-time system.
Difference equations are easily rearranged to obtain recursive formulas for comput
ing the current output of the system from the input signal and past outputs. We rewrite
Eq. (2.36) so that y[n]is alone on the left-hand side:
M N
1 1
y[n] =bx[n - k] - 44y[n - k].
do k=1

This equation indicates how to obtain y[n] from the present and past values of the input
and the past values of the output. Such equations are often used to implement discrete-time
systems in a computer. Consider computing y[n} for n > 0 from x[n] for the second-order
difference equation (2.37), rewritten in the form
1
y[n] = x[n) + 2x[n- 1]-y[n - 1)-yln-
4 2]. (2.38)

Beginning with n = 0, we may determine the output by evaluating the sequence of equations
y[0] = x[0] + 2x[-1] - y[-1] -yl-2], (2.39)
1
y[1] =x[1] + 2x{0] - y[0] - -1), (2.40)

y[2] = *[2] + 2x[1] - y[1]-o],

y[3] = x(3] + 2x(2] - y[2] - yl1 ),

In each equation, the current output is computed from the input and past values of the out
put. In order to begin this process at time n = 0, we must know the two most recent past
values of the output, namely, y[-1)and y[ -2). These values are known as initial conditions.
2.9 Diferential and Difference Equation Representations of LTI Systems 143

The initial conditions summarize all the information about the system's past that is
needed to determine future outputs. No additional information about the past output is nec
esary. Note that, in general, the number of initial conditions required to determine the
output is equal to the maximum memory of the system. It is common tochoose n = 0or
t= 0as the starting time for solving adifference or differential equation, respectively. In
this case, the initial conditions for an Nth-order difference equation are the Nvalues
y[-N], y[-N+ 1],...yl-1],
and the initial conditions for an Nth-order differential equation are the values of the first
N derivatives of the output-that is,
d
y(t)l,-oy()lo'd2) =0 =0

directly related
The initial conditions in adifferential-equation description of an LTIsystemare voltages on ca
to the initial values of the energy storage devices in the system, such as initialthe initial condi
discrete-time case,
pacitors and initial currents through inductors. As in the
can affect future
tions summarize all information about the past history of the system that systems.
continuous-time
outputs. Hence, initial conditions also represent the "memory" of
EQUATION Find the first
EXAMPLE 2.15 RECURSIVE EVALUATION OF ADIFFERENCE
by Eq. (2.38), assuming that the
two output values y[0]and y1]for the system described y[-1] = land y[-2] = -2.
input is x[n]= (1/2)'"u[n]and the initial conditions are
to obtain
Solution: Substitute the appropriate values into Eq. (2.39)
1
y[0] = 1 +2 X0 -1 -x(-2) =;
Now substitute for y[0] in Eq. (2.40) to find
1 3
(1] =;+2 x1- x (1) =1
COMPUTER
DIFFERENCE EQUATION BY MEANS OF A
EXAMPLE 2.16 EVALUATION OF A
difference equation
Asystem isdescribed by the
2] =
y[n] - 1.143y[n - 1] + 0.4128yl[n - 0.0675x(n] + 0.1349x[n - 1] + 0.675x[n - 2).
present output from the past outputs and the
Write a recursive formula that computes the system
the step response of the system, the
current inputs. Use a computer to determine conditions are y-1]= 1and yí -2] = 2.,
output when the input is zero and the initial
inputs *[n = cos on), x[n] = cosn),
and the output in response to the sinusoidal conditions. Last, find the output of the system
and xn] = cos1on), assuming zero initial stock depicted in Fig. 2.27, assuming zero
if the input is the weekly closing price of Intel
initial conditions.

Solution: We rewrite the difference equation as

y[n] = 1.143y[n - 1) - 0.4128y[n - 2]
+ 0.0675x[n] + 0.1349x[n - 1) + 0.675x{n - 2].
 147
2.J0 Soling Differential and Difference Equations

2.10 Soving Differential

4nd DifferenceEquations
differential and difference equa
In this section, we briefly review one method for solving solutions that provides insight into
tions. Our analysis offers a general characterization of
LTIsystem behavior.
or difference equation may be ex
The output of a system described by a differentialof the homogeneous form of the dit
solution
pressed as the sum of two components. One is a
the bomogeneous solution and denote by
ferential or difference equation, which we shall ternmoriginalequation, which we shall term the
the
y". The second component is any solution ofthe complete solution is y = yl" + ylP. (Note
particular solution and denote by yP), Thus, referring to both continuous and discrete
time.)
when
that we omit the arguments (t) or [nl

2.10.1 THE HoMOGENEOUS SoLUTION

setting all
form of a differential or difference equation is obtained by
The homogeneous
zero. Hence, for a continuous-time system, y"() is the solu
termsinvolving the input to
tion of the homogeneous equation

k=0
a dze() = 0.
is of the form
homogeneous solution for acontinuous-time system
The
N
(2.41)

the system's characteristic equation

where the r,are the N roots of
N (2.42)
= 0.

equation establishes the fact that y()

Substitution of Eq. (2.41)into the homogeneous
set of constantsc,. homogeneous equation
is a solution for any solution of the
Indiscrete time, the N
Say"[n - k] = 0
k=0

is N
(2.43)
=

system's characteristic equation

roots of the discrete-time
where the r,are the N
(2.44)
a,Nt = 0.
that
the homogeneous equation establishes the fact
(2.43) into later, in
Again, substitution of Eq. set of constanis c,. In both cases, the c, are determined and
yl[n] is a solution for any satisfy the initial conditions. Note that the continuous-
solution
order that the complete equations are different.
discrete-time characteristic
148 CHAPTER 2 TIME-DoMAIN REPRESENTATIONs OF LINEAR TIME-INVARIANT SrsTEMS
R

x(t) C y(t)

FIGURE 2.30 RC circuit.

The form of the homogeneous solution changes slightly when the characteristic eaua
tion described by Eq. (2.42) or Eq. (2.44) has repeated roots. If aroot r, is repeated ptime:
then there are p distinct terms in the solution of Eqs. (2.41) and (2.43) associated with r
These terms respectively involve the p furctions
e'', te'',... , ;P-'e}
and
r, nrf,... ,n'.
The nature of each term in the homogeneous solution depends on whether the roots
r; are real, imaginary, or [Link] roots lead to real exponentials,imaginary roots to
sinusoids, and complex roots toexponentially damped sinusoids.

ExAMPLE 2.17 RC CIRCUIT: HoMoGENEOUS SoLUTION The RC circuit depicted in

Fig. 2.30 is described by the differential equation
d
y(t) + RC,y() = x(1).
Determine the homogeneous solution of this equation.
Solution: The homogeneous equation is
v() + RC
d() = 0.
The solution is given by Eq. (2.41), using N = 1 to obtain
yl() = cGeiv,
where r, is the root of the characteristic equation
1 + RCr, = 0.
Hence, r = RC> and the homogeneous solution for this system is
y"(t) = Gje RC V.

EXAMPLE 2.18 FIRST-ORDER RECURSIVE SYSTEM: HOMOGENEOUS SoLUTION Find

the homogeneous solution for the first-order recursive system described by the differ
ence equation
y[n] - pyln - 1] = x[n].
Solution: The homogeneous equation is
y[n] - py[n - 1] = 0,
and its solution is given by Eq. (2.43) for N = 1:
2. 10 Solving Differential and Difference Equatios 149

Theparameter r, is obtained from the root of the characteristic equation given by q. (2*)
with N= 1:
p 0,
Hence, r,p, and the homogeneous solution is

described by
Problem 2.16 Determine the homogeneous solution for the systems
the following differential or difference equations:
(a)
d|Y(1) + Sy0) + 6y(:) = 2x(t) +
(b) d
d
d2V)+ 3 )+2y(t) = x(t) +
(c)
y[n] - (9/16)y[n - 2] = x[n - 1]
(d)
y[n]+ (1/4)y[n - 2] =x(n] + 2x(n - 2]
Answers:
(a)
y(:) = Ge+ Ge
(b) y"(t) = Ge t e

(c)
yl[n]= G(3/4)" + c(-3/4)"
(d)
yln] = (1/2e"2)" + (1/2e m2y"
depicted
Determine the homogeneous solution for the RLC circuit
º Problem 2.17 and C so that
of R, L, and C. Indicate the conditions on R, L,
in Fig. 2.26 as a function sinusoids, and exponen
consists of real exponentials, complex
the homogeneous solution
tially damped sinusoids.
Answers: For R?
y"(t) = Ge'i t e',
where
-R + VR' -R - VR' -
and = 2L
2L

For R²= 1
yl"(1) = ce 21' + c,te
exponentials for R' Ccomplex sinusoids for R = 0, and
The solution consists of real
exponentially damped sinusoids for R' s .
150
CHAPTER 2 m TIME-DoMAIN REPRESENTATIONS OF LINEAR TIME-INVARJANT SYsTEMS
TABLE 2.3 Form of Particular Solutions Corresponding to Commonly
Used Inputs.
Continuous Time Discrete Time
Particular
Input Solution
Particular
lnput Solution
1
1
t
C,t t C
e at ce
-at
a" ca"
cos(wt t+ ) C cos(wt) + c sin(wt) cos(Rn + ) G, cos(n) + c; sin(ln)
s 2.10.2 THE PARTICULAR SOLUTION
The particular solution y) represents any solution of the
for the given input. Thus, yP) is not differential or difference equation
unique.
suming an output of thesame general form as the A particular solution is usually
obtained by as
crete-time system is x[n) = a", then we assume that the input. For example, if the input to a dis
and find the constant cso that yP[n] output is of the form yPn] = ca"
is a
input is x[n]= A cos(n + ), then we solution of the system's difference equation. If the
yP'[n] cos(n) + c sin(n), whereassume c and
a general sinusoidal
response of the
c, are determined so that yP[n] form
the system's difference equation. satisfies
sistent with our expectation that Assuming an output of the same form as
the output of the system be the input is con
This approach for finding a particular directly related to the input.
form as one of the components of the solution is modified when the input is of the same
aparticular solution that is homogeneous solution. In that case, we must assume
independent of all terms in the
accomplished analogously to the procedure for generating homogeneous solution. This is
components when there are repeated roots in the
tiply the form of the particular characteristic independent natural-response
sponse component not included insolution by the lowest powerequation.
of
Specifically, we mul
t or n that will give a re
by substituting the assumed the natural response, and then we
solve for the coefficient
particular solution into the differential or
The forms of the particular
in Table 2.3. More solutions associated with common input difference equation.
extensive tables are given in books
ferential equations, such as those listed under signals are given
devoted to solving difference
The particular solutions given in Further Reading at the end of this and dif
Table 2..3 assume that the inputs
input is specified after a starting time exist for
chapter.
t = 0 or n all time. If the
in solving
differential or difference equations = 0(e.g., x(t) = e u(t)], as is common
ular solution is valid only for t > 0 subject to initial
or n >0. conditions, then the partic
EXAMPLE 2.19 FIRST-ORDER RECURSIVE
Find a particular solution for the SYSTEM (CONTINUED): PARTICULAR
equation first-order recursive system described by the SOLUTION
difference
ifthe input is x{n] =
y[n] - py[n - 1] = x[n]
(1/2)".
Solution: We
yP[n] and x[n) assume
a particular solution of
the form
into the given difference equation yields yP[n] = )". Substituting

(4)-)-9):
2.10 Solving Differential and Difference 151
Equations
We multiply bothsides of the equation by (1/2)" to
obtain
(2.45)
c(1 - 2p) = 1.
Solving this equation for c, gives the particular solution
1
ylPl[n] =1 - 2p
Ifp =(), then the particular solutioD has the same form as the homogeneous solution
found in Example 2.18. Note that in this case no coefficient c, satisfies Eq. (2.45), and We
must assume a particular solution of the form yl[] =G,n(1/2)". Substituting this par
ticular solution into the difference equation gives con(1 - 2p) + 2pc, = 1. U1ng
p= (1/2) we find that c, = 1.

ExAMPLE 2.20 RCCRCUIT (CONTINUED):PARTICULAR SOLUTION Consider the KC

circuit of Example 2.17 and depicted in Fig.: 2.30. Find a particular solution for this sys
tem with an input x(t) = cos(agt).
Solution: From Example 2.17, the differential equation describing the system is
d
y(:) +RCy(t) =x().
cos(wt) + c sin(wt). Replacing
We assume a particular solution of the form ylP(t) =
y(t) in the differential equation by ylP(t) and x(t) by cos(wt)gives
= cos(w).
Gcos(@t) + sin(wt) - RCwnc, sin(wt) + RCwc, cos(wt)
coefficients of cos(wpt)
The coefficients c, and c, are obtained by separately equating the
in two unknowns:
and sin(wot). This gives the following system of two equations
G t RCwc, = 1;
-RCwnc t = 0.
Solving these equations for c and c, gives
1
C =
1+ (RCw)'
and
RCwo
1+ (RCw)
Hence, the particular solution is
1 RCoo
Cos(@) + 1+ (RCw)? sin(wt) V.
1+ (RCu )2

º Problem 2.I8 Determine the particular solution associated with the specified input
differential or difference equations:
for the systems described by the following
(a) x(t) = e:
d
dy2(4)+ s di) + 6y(r) = 2x() + d ()

(b) x(t) = cos(2t):

d d
2Y() +3 ( ) + 2y(:) = x(1) +
CHAPTER 2 TIME-DoMAIN REPRESENTATIONS OF LINEAR TIME-INVARIANT SYsTEMS

(c) x{n] = 2:
vn] - (9/16)y(n - 2] =xn- 1]
(d) x(n) = (1/2)":
y[n] + (1/4)y[n - 2] = x[n] + 2x(n - 2]
Answers:
(a) y(t) = (1/2)e
(b) yP (1) = (1/4) cos(2:) + (1/4) sin(2r)
(c) yPn] = 32/7
(d) yP[n] = (9/2)(1/2)"
2.10.3 THE CoMPLETE SOLUTION
The complete solution of the differential or difference equation is obtained by
particular solution and the homogeneous solution and finding the unspecifiedsumming
the
coefficients
in the homogeneous solution so that the complete solution satisfies the prescribed initial
conditions. This procedure is summarized as follows:

Procedure 2.3: Solvinga Diferential or Difference Equation

1. Find the form of the homogeneous solution yl") from the roots of the
characteris
tic equation.
2. Find a particular solution yP) by assuming that it is of the same form as the input,
yet is independent of all terms in the homogeneous solution.
3. Determine the coefficients in the homogeneous solution so that the complete so
lution y = yl) + y satisfies the initial conditions.

We assume that the input is applied at time t 0or n = 0, so the particular solution
applies only to times t > 0or n > 0, respectively. If so, then the complete solution is valid
only for just those times. Therefore, in the discrete-time case, the initial conditions y[ -N],..,
yí-1] must be translated to new initial conditions y[0],..., y[N - 1] before Step 3 is per
formed. Translation of the initial conditions is accomplished by using the recursive form of
the difference equation, as shown in Example 2.15 and in the examples that follow.
In the continuous-time case, the initial conditions at t= 0 must be translated
to t = 0* to reflect the effect of applying the input at t= 0. While this process is often
straightforward in problems involving capacitors and inductors, translating initial
conditions for the most generaldifferential equation is complicated and will not be dis
cussed further. Rather, we shall only solve differential equations for which application
of the input at t = 0 does not cause discontinuities in the initial conditions.À neces
sary and sufficient condition for the initial conditions at t = 0* to equal the initial
conditions at t =0 for a given input is that the right-hand side of the differential
equation in Eq. (2.35), S-0bx(t),contain no impulses or derivatives of impulses.
For example, if M = 0, then the initialconditions do not need to be translated as long
as there are no impulses in x(1), but if M= 1, then any input involving a step dis
continuity at t = 0generates an impulse term due to the x(t) term on the right-hand
side of the differential equation, and the initial conditions at t = 0* are no longer
equal to the initial conditions at = 0. The Laplace transform method, described in
Chapter 6, circumvents these difficulties.
 10 Solving Differential and Difference Equations 153

EXAMPLE 2.21 FIRST-0ORDER RECURSIVE SYSTEM (cONTINUED): CoMPLETE SOLUTION

Find the solution tor the first-order recursive system describedby the difterence equation
1
(2.46)
y[r) -ln - 1] =x(n)
if the input is x(n] = (1/2)"u[n] and the initial condition is y[-1] 8.
Solution: The form of the solution is obtained by summing the homogeneous solution
determined in Example 2.18 with the particular solution determined in Example 2.19 after
setting p = 1/4:

for n > 0. (2.47)

condition. First, we translate the initial condi

The coefficient c is obtained from the initial obtain
to time n = 0by rewriting Eq. (2.46) in recursive form and substituting n = 0 to
tion
y[0] = *(] + (1/4)y[-1],
3 into
implies that y[0] = 1+ (1/4) X 8 = 3. Then we substitute y[0]=
which
Eq. (2.47), yielding

complete solution as
from which we find that c = [Link], we may write the
for n > 0.

RESPONSE Find the

complete
(CONTINUED): CoMPLETE assum
EXAMPLE 2.22 RC CIRCUIT Fig. 2.30to an input x(t) = cos(t)u(t)V, across
depicted in initial voltage
response of the RC circuit 1N and C= 1F andassuming that the
R=
ing normalized values
= 2 V.
the capacitor is y(0) solution was obtained in Example
2.17:
homogeneous
Solution: The
yl^() = ceRCV.
input in Example 2.20, namely.
solution was obtained for this
Aparticular RC
1 cos(t) + 1+ (RC) sin(t) V,
1: +(RC)?
SübstitutingR = 1 and
C= 1E, we findthat the complete
wn = 1.
where we have used
solution is
1 t>0.
y(t) = ce t jcos +;sinV
154 CHAPTER 2 TiME-DoMr RE PRESENTATIONS OF LINEAR TIME-IwARIANT S1sTEMS

The input does not introduce impulses into the right-hand side of the different1al eQuas,
so the coefficient c is determined from the initial condition y(0) =y0) = 2. We ha
1 1
2 = ce cos0° - 2
1

so that c = 3/2, which gives

1
y)=e 2
t>0.

EXAMPLE 2.23 FINANCIAL CoMPUTATIONS: LOAN REPAYMENT Example 2.5 showe

that the first-order difference equation introduced in Section 1.10 and studied in
ples 2.18, 2.19, and 2.21 may be used to describe the value of Exam.
an investment earning a
fixed rate of interest. The same equation also describes the balance of a loan if
xn'<i
represents the principal and interest payment made at the beginning of each period and y'n
is the balance after the principal and interest payment is credited. As
before, if % is the
interest rate per period, thenp = 1+r/100.
Use the complete response of the first-order difference equation to find
the
required to pay off a $20,000 loar in 10 periods. Assume equal payments and a paymen:
10% in
terest rate.

Solution: We have p = 1.1 and y-1] = 20,000, and we assume that xn = b is the
payment each period. Note that the first payment is made when n = 0. Since the loan bal
ance is to be zero atter 10payments, we seek the payment b for which y 9 = 0.
The homogeneous solution is of the form

y"n)= G(1.1)",
while the particular solution is of the form

Since the input (the payment) is constant. Solving tor c, by substituting y ? n = c, and
xn = bintothe difference equation y[n - 1.1yn - 1] =xn, we obtain
G, =-10b.

Therefore, the complete solurion is of the form

y[n) = c,{1.1)" - 10b, n0, (2.48)
We solve for c, by first translating the initial condition forward one period to obtain

y(0) = 1.1y[-1] + x{0)

= 22,000 + b.
Next, we substitute y( 0} into Eq. (2.48) to obtain the equation for ch:

22,000 + b= {1.1y - 10b.

2.10 Solving Differential and Difference Equations 155

20,0009
18,000

l6,000

14,000

12,000
yln)
10,000
8000

6000
4000

2000

) 6 7 8

FIGURE 2.31 Balance on a $20,000 loan for Example 2.23 in U.S. dollars. Assuming 10% inter
est per period, the loan is paid off with 10 payments of $3,254.9 1.
Thus,c, = 22,000 + 11b. This implies that the solution of the ditference equation is given by

y[n] = (22,000 + 11b)(1.1)" - 10b.

We now solve for the required payment bby setting y| 9 = 0. That 1S,
0= (22,000 + 11b) (1.1)' - 10b,
which implies that
-22,000(1.1)
h=
11(1.1)'- 10
= -3,254.91.
in 10 payments.
Hence, a payment of $3,254.91 each period is required topay off the loan
Figure 2.31 depicts the loan balance y[n ].
conditions, for the systems
º Problem 2.19 Find the output, given the input and initial
described by the following differential or difference equarions:
1d
(a) x(1) =e'u(t), y(0) =
)t6y(1) =x(1)
4 d 3
(b) x() = cos(t)u(), y(0) =

deY() + 2y(?) = 2x(t)

(c) x(nj= u[n], yl-2] = 8, y[- 1] =0:
1
y[n]y[n - 2) =2x[) +x<n - 1]
(d) x(n] 2"uin), y[-2] = 26, yl-1] = -1:
st) -(Jin-)-(r- 2) -snl+(-1)
 TIME-INVARIANT SYSTEMs
REPRESENTATIONS OF LINEAR
TIME-DoMAIN
CHAPTER2
156

Answers:

Juo)
t) - ((GJe+eY-2e
(a)

(b)
sin(t) - 2e + e

(c)

(d)

following
Problem 2.20 Find the response of the RL circuit depicted in Fig. 2.29 to the A:
º current through the inductor is y(0) = -1
inputvoltages, assuming that the initial
(a) x(t) = u(t)
(b) x(t) = tu(t)
Answers:
(a)
Y) -(-(1+ ) A, t>0

(b)
L
t>0

2.11 Characteristics of Systems Described

by Differential and Difference Equations
It is informative to express the output of a system described by a differential or difference
equation as the sum of two components: one associated only with the initial conditions, the
other due only to the input signal. We willterm the component of the output associated
with the initial conditions the natural response of the system and denote it as y "), The
component of the outputdue only to the iaput is termed the forced response of the system
and is denoted as y), Thus, the complete output is y = y) + y),
2.11.1 THE NATURAL RESPONSE
The natural response is the system output for zero input and thus describes the manner
in which the system disipates any stored energy or memory of the past represented by
non-zero initial conditions. Since the natural response assumes zero input, it is obtained
from the homogeneous solution given in Eq. (2.41) or Eq. (2.43) by choosing the coeffi
cients c, so that the initial conditions are satisfied. The nYtural response assumes zero
input and thus does not involve a particular solution. Since the homogeneous solutions
apply for alltime, the naturalresponse is determined without translating initialcondi
tions forward in time.
 2.1| Characteristics of Systems Described by Differential and Difference 157
Equations
EXAMPLE 2.24 RC CIRCUIT (CONIINUED): NATURAL RESPONSE The system in Ex
amples 2.17, 2.20, and 2.22 is described by the differential equation
d
y(1) + RC dt
v(t) = x(t).
Find the natural response of this system, assuming that v(0) = 2 V,R = 12, and C= I.
Solution: The homogeneous solution, derived in Example 2.17, is
y(1) = c,eV.
Hence, the naturalresponse is obtained bychoosingG, so that the initial condition y""(0) 2
is satisfied. The initialcondition implies that c, = 2, so the natural response is
y"(t) = 2eVfor t> 0.

EXAMPLE 2.25 FIRST-ORDER RECURSIVE SYSTEM (CONTINUED): NATURAL RESPONSE

The system in Example 2.21 is described by the difference equation
1
y[n] -yn - 1] =x(n],.
Fird the natural response of this system.
Solutio: Recallfrom Example 2.21 that the homogeneous solution is

Satisfaction of the initial condition y[-1]= 8 implies that

or c, = 2. Thus, the natural response is

n >-1.

described by the
Problem 2.21 Determine the natural response for the systems
the specified initial conditions:
following differential or difference equations and
d -7:
(a) y(0) =3, dyDo=
d d
dz(1)+ )+ 6y(t) = 2x(t) + d()
d
(b) y(0) = 0, ()eo = -1: d
z(1) + 3,y(0) + 2y(1) = x(1) + d*()
16/3:
(c) y[-1)= -4/3, y[-2] = (9/16)y[m - 2] =x(n- 1]
y[n] -
(d) y[0)= 2, y[1] = 0:
y[n] + (1/4)y[n - 2] = x{n] + 2x[n - 2]
158 CHAPIER 2 TIME-DoMAIN REPRESENTATIONS OF LINEAR TIME-INVARIANT SYSTEMS

Answers:
(a)
y'"'(t) =e+ 2e 2 for t>0
(b)
y"(t) - e t e , for t >0
(c)
y"[n] = (3/4)" + 2(-3/4)", for n >-2
(d)
y"[n] = (1/2e/2y" + (1/2e im/2 )", for n>0
2.11.2 THE FORCED RESPONSE
The forced response is the system output due to theinput signal assuming zero initialcon:
ditions. Thus, the forced response is of the same form as the complete solution of the dif.
ferential or difference equation. A system with zero initial conditions is said to be "at rest,"
since there is no stored energy or memory in the system. The forced response describes the
system behavior that is *forced" by the input when the system is at rest.
The forced response depends on the particular solution, which is valid only for
times t >0 or n >0. Accordingly, the at-rest conditions for a discrete-time system,
y[-N] = 0,...,yl-1] = 0, must be translated forward to times n = 0, 1,... , N- 1be
fore solving for the undetermined coefficients, such as when one is determining the com
plete solution. As before, we shall consider finding the forced response only for
continuous-time systems and inputs that do not result in impulses on the right-hand side
of the differential equation. This ensures that the initial conditions at t =0* are equal to
the zero initial conditions at t = 0.

ExAMPLE 2.26 FIRST-ORDER REÇURSIVE SYSTEM (cONTINUED): FoRCED RESPONSE

The system in Example 2.21 is described by the first-order difference equation
1
y[n]-y[n - 1] = x[n].
Find the forced response of this system if the input is x[n] (1/2)"u[n].
Solution: The difference between this example and Example 2.21 is the initial condi
tion. Recall that the complete solution is of the form
n>0.

To obtain G,we translate the at-rest condition y[ -1] =0 to timen = 0 by noting that
1
y[0] = x(0] +-1),
which implies that y[0] = 1 + (1/4) X 0. Now we use y[0] = 1to solve for c, from the
equation
2.11 Characteristics of Systems Described by Differential and Diference Equations 159

which implies that = [Link], the forced response of the system is

n0.

EXAMPLE 2.27 RCCIRCUIT (CONIINUED): FoRCED RESPONSE The system in ta

ples 2.17, 2.20, and 2.22 is described by the differential equation

Y) + RC )=x().

and
riltd tne Orced response of this system, assuming that x(t) = cos(t)u(t) V, R = 12,
C= 1E

Solution: Example 2.22 established that the complete response isof the torm
1 1
y(t) = ce + coS t + Sin t V, t>0.
2

is ini
The forced response is obtained by choosing c under the assumption that the system
-1/2, and
obtain c =
tially at rest-that is, assuming that y(0) = y(0*) = 0. Thus, we
the forced response is given by
1 .. 1 1
y() = -;e,costsinntV.
t
naturalresponse in Example 2.24 is equal
Note that the sum of the forced response and the
Example 2.22.
to the complete system response determined in
the systems described by the fol
Problem 2.22 Determine the forced response for
specified inputs:
lowing differential or difference equations and the
(a) x(t) = eu(t)
y() + sy) + 6y(:) = x(1)
(b) x(t) = sin(2t)u(1)
d
d2)+ 3d
) + 2y(1) = x(t) -

(c) x[n] = 2u[n)

y[n] - (9/16)y[n - 2] = x[n - 1]
Answers:
(1/2)e)u(:)
(a) y() = (1/2)e- e+
+ (1/4)e)u(t))
(b) y(t) =((-1/4) cos(2) + (1/4) sin(21)
(4/7)(-3/4)")u[n]
(c) yn) = (32/7 - 4(3/4)" -