Introduction to Systems 2.

17

Practice Problemn 2.3

Check whether the following systems are time-invariant.

ö) y() = x{t)dr
-00
Ans: Time-invariant

(i) dy()
dt
+5ty(r) =x() Ans: Time-variant

(ii) y() =x(t+ 10)+x*() Ans: Time-invariant

(iv) y(n)=x(-n) Ans: Time-variant

(v)'y(n) =x(n+ 1) +x(n) +x(n-1) Ans: Time-invariant

(vi) y(n) = sinlr(n)) Ans: Time-invariant

2.2.7 Stable and unstable systems

x(0)
x)E
bounded bounded
Isignal signal
(within the (within
limits)
limits)

(a) (b)

x4()

unbounded unbounded
signal signal

(c). (4)

Fig. 2.5 Examples for bounded and unbounded slgnals.

 &K ii the condition r
S tsilisbnd if it stisfes
tbe
Fi unbounded signals are coo

Fg,25 Typicalstable and unstable systems

A nssilDbe
bocedtndbounded output stable ifand r5t
ybniadtprotasabounded outgut. Fig26shows typical stable at

iwgprDbT weher a systmi stzble or not This an be

i a e beie boaedge cfipase rsponse of the system Tbeefei
Seed saaspic sl be covered in the forth coning
chapters

Sotred Problera 26
Check sbeter te fowg systens are
() Saticr ie
)b r a
(c) CzslraOGL
(einzitrtine-rit
2-s0)=x)
09,0+70-)
 Introduction to Systems 2.19

(iv) y(n) =x)x(n-1)

(v) y(n) = cos x(n))
(vi) y(n) = log1o |x(n)|
n+1
(vii) y(n) =k=-o x(k) od

(vii) y(n) = sgnlx(n))

(ix) y(n)=x(n)u(n)
i) ) Given y)043)+y)
di2 dt
=x()
(a) The system is described by differential equation. Hence it is a dy
namic
(b) The system.
input x1() and output y1(t) can be described by the differential
equation as

yi()143
di2
+3 n4y() =x)
dt
(2.44)

Similarly for input x2(1) and output y2(t) the differential equation can
be written as
H+32)+n() =x(t) (2.45)
dt2 dt
Multiply Eq. (2.44) with a and Eq. (2.45) with b and add. Then we
get
+3ar1(e)tan()+by()2)
d2 dt dt2
+362()
dt
+byz() = ax1 (1) + bx2 ()

3dt lv()+ by2())

tay1() +byz(1) =ax1(t) +bxz() (2.46)
Shaded part of the Eq. (2.46) is not afunction of weight sum of out
puts. Hence superposition principle is not satisfied, and the system is
non-linear.
(c) The output depends on the present input only. Hence the system is
causal.
2.20 Signals and Systems
(d) The coefficients ofthe diferential equation are function of time.
br
the system is time-variant. Hence
causal and time variant.
The system is dynamic, non-linear,
ii) Pyl)4dy)
dt3 dt2
+ Sdy)+240)=x()
+
dt
(a) The system is described by a differential equation. Hence it is a dy.

(b) namic system.

For input x1 (), let the output be yn(). Then the differential equation
can be written as
dnl)
d3
4yn)
+
di2
5dn)+20)
dt
=2)
+2yi() =n() (2.4T)
Similarly for an input x2(), let the output be y2(1). Then the differen.
tial equation can be written as

d3 dt2
+ dt
+23) =x(t) (248)
The linear combination of Eq. (2.47) and Eq. (2.48) can be written as
Pn)dn)n()
d3 d2
+2ayi()+b )
dt dt3
+462()
di? dt +2by() = axn(t) +bxnl)
on)+byn())lgy()
d3 dt2
+byal)dlon)
+5
dt
+bya))
42oitO2bi)ar () +bx2() (2.49)
Shaded portion in the LHS is function of weighted sum of
outputs. Superposition principle is not satisfied. squares of
is non- linear. Therefore the system
(C) The output depends on present
causal, input only. Therefore, the system
(d) The coefficient of the
system is time-invariant.differential equation are constants. Hence the
The system is dynamic,
i) Given non-linear, causal andtime-invariant
) =od{xtO}
(2.50)
 Introduction to Systems 2.21

(a) For any value of t the output depends on the stored value of input.
Hence the system is dynamic.
(b) For an input x1()
n)= Tn(0))=() -n(-)) (251)
For an input x2(1)
1
y2() =T|a())=)-x(-) (2.52)

The output due to weighted sum of inputs is

ys) =Tlan (0)+ bra(0)] =;la () +bra() (ax(-1)

+ br(-t))] (2.53)
The weight sum of outputs is

an() +bya() =~a1(0) -(-)]l+;bia(0) -a(-O) 2s4)

ys(4) = ay1 () +by2()
Eq (2.53) and Eq (2.54) are equal. Therefore, the system is linear.
(c) For negative values of t the output depends on future values of input.
Hence the system is non-causal.
(d) Given
y) =Tk0))=()-x(-))
The output due to delayed input is'

y,T) =Iz(t-)]=;ke-)-x-) (2.55)

The delayed output is

t-)=z-T)--(-T)
1
=k(-)-x{-t+T) (2.56)
y(t, T) #y(t-T)
Hence the system is time-variant.
The system is dynamic, linear, non-causal and time-variant.
2.22 Signals and Systems

iv) y(n) =x(n)x(n- 1)

It requires
(a) The output depends on past values of input.
Therefore, the system is dynamic. menory.
(b) The output y1 (n) due to input x1 (n) is
yi(n) =x(n)x (n- 1) (251,
For an input xz (n)
y2(n) =x2(n)r2(n-1) (2.5,
Theoutput due to weighted sum of
inputs is

ys{n) = Tlar1 (n) + bx2(n)]= (axj (n) + bxz(n)] (ax1 (n - 1)

+bxz(n1)) (2.59
The weighted sum of outputs is given by
ay1 (n) +by(n) =ar (n)xi(n-1) + bx2 (n)x2(n1)
B(n). ay (n) + by2 (n) (2.60
Eq(2.59) and Eq (2.60) are not equal. Hence the system is non-linex
(c) The output depends on present and past values of the input. Hence te
system is causal.
() Given y(n) = Tk(n)] = x(n)x(n -1)
The output due to delayed input is

y(n, k) = Tr(n)] =x(n-k)x(n-1-k) (2.61)

The delayed output

(26)
y(n-k) =x(n-k)x(n -1-)
y(n,k) =y(n-k)
Therefore, the system is time-invariant.
The system is dynamic, non-linear,
v) y(n) =coss*(n)) causal and time-invariant.
(a) The output at any instant depends on the input at that instant. Ther
fore, the system is statíc.
 Introduction to Systems 2.23

(b) For an input x1 (n)

yn(n) = cosr1(n)) (2.63)
For an input x2(n)
a(n) =cos<2(n) (2.64)
The output due to weighted sum of inputs is

ys(n) = T(a,x (n) +bx2(n))

o = cos(a1x1(n) + bxz (n)] (2.65)

The weighted sum of outputs is given by

ayn (n) + by2(n) = acos*1 (n)) + bcosk2(n)) (2.66)

ys(n) ay (n) + by2(n)
Eq(2.65) and Eq (2.66) are not equal. Hence the system is non-linear.
(c) The output depends on present inputs only. Hence the system is causal.
() Given y(n) = Tk()]= coslx(n)]
The output due to delayed input
y(n,) = Tl(n- )]=coslx{n-* (2.67)

The delayed output is

y(n-) = cos x(n- )], (2.68)
y(n,k) =y(n-)
Therefore, the system is time-invariant.
The system is státic, non-linear, causal and time-invariant.
vi) Given y() =log1o l=(n)l,
(a) The output at any instant depends on the input at that instant. Hence
the system is static.
(b) Given y(n) = log1o x(n)|
For input xj(n)
y1(n) = logjo ki (n)l (2.69)

4 For input x2(n)

n(n) = logjo a(n)| (2.70)
2.24 Signals and Systems

The output due to weighted sum of inputs is

s(n) =Tlax1 (n) +bx(n)] = log1olax1 (n) + bx (n)| (2.11)

The weighted sum of outputs
l*2(n)|
ayn (n) + hy2(n) = alogjo lx1(n)|+blog1o (2.72)
ys(n) ayi(n) +byz(n)
Eq(2.71) and Eq (2.72) are not equal. Hence the system /s non-line.
system is calusal
(c) The output depends on present input only. Hence the
() y(n) = Tk(n)]= log o lx(n)|
The output due to delayed input

y(n, k) = Tk(n)] = logo lx(n- k)| (2.73)

The delayed output
y(nk) = logo Ix(n)| (2.74)
’ y(n, k) =y(n- k)
Hence, the system is time-invariant.
The system is static, non-linear, causal and tim-invariant.
n+1
vi) y(r) =E x(4)
(a) The output

yn)=x-o) +......+x(0) +x(1) +...x(n) +x(n+ 1) for n>0

The output depends on stored values of input. Hence the syste
dynamic.
(b) For inpyt x1(n)
n+1
y1(n) = 1(k)o (2
k<o
For input x(n)
n+1
(2.70
n(n)= (0)
 Introduction to Systems 2.25

The input due to weighted sum of inputs is

n+I
(2.77)
ys(n) =T|ax1 (n) +bxz(n)] = k=-o lax (k) + bx2(k)
The weighted sum of outputs is
n+1 n+1

ay1(n) + by2(n) = a n(k)+b n(k)

k=-o k=-o
n+1 n+1
=
ax (*) + bra(k) (2.78)
k=-0 k=-oo

ys(n)= ayi(n) +by2(n)

Eq (2.77) and Eq (2.78) are equal. Hence he system is linea.
n+1
(c) y(n) =k=-oo x(k)t So
For n=0

y(0) =k=-o
| ) =xf-o)+...+ x(0) +x(1)
past inputs present future
input input
The output depends on future value of input. Hence the system is
noncausal.
n+1
vü) y(n) =Tz(n)] =k=-ox(*)
The output due to shifted input
n+1-k
y(n, ki) = Tx(7-ki)l= k=-o
x(k) (2.79)

The delayed output is n+1

y(n k) =k=loox(k-ki) (2.80)

nt1-kgg
k=-oo

y(n,ki) =y(n -k)

Hence the system is time-invariant.
2.26 Signals and Systems

ix) y(n) = sgn(n)]

sgnx(n)]= lforn >0
=-1forn <0

1 2 3 4 n

Fig. 2.7 Signum function.

(a) The output depends on the input at that instant only. Therefore the
system is static.
(b) For an input x1(n)
y1(n) =Tki(n)]= sg1(n)] (2.81)
For an input x2(n)
y2(n) = T2(n)] = sgnla(n)] (2.82)
The output due to weighted sum of inputs is

»(n) =Tlax| (n) +bxz(n)]= sgnlax1(n) + bxa(n)) (2.83)

The weighted sum of outputs is
ay1(n) +byn(n) = asgnl1(n)] + bsgn2()] (2.84)
ys(n) # ayi(n) + by(n).
Eq (2.83) and Eq (2.84) are not
linear. equal. Therefore, the system is no
(c) The output y(n) depends on
present input. Hence the system: is causal.
(d) Given y(n) =sgn(n))
The output due to delayed input
y(n,k) =Tk(n )] = sgnlr(n -)] (285)
 Introduction to Systems 2.27

The delayed output

y(n-k) = sgnx(n -k) (2.86)
y(n, k) =y(n-k).
Hence the system is time-invariant.
x) Given yln) =x(n)u(n)
(a) The output depends on present input only. The system won't require
ny memory. Hence the systém is static.
(b) For input x1 (n)
yi(n) = Tl1()] =x1(n)u(n) (2.87)

For input x2(n)

n(n)= T[n()] =x2(n)u(n) (2.88)

The output due to weighted sum of inputs

vlo
ys(n)=T[ax| (n) +brn(n)]= (ax1(n) +bxn(n)]u(n) (2.89)

The weighted süm of outputs is

ay1 (n) + byz(n) = ax(n)u(n) + bxz(n)u(n)
- lax (n) +bxs(n)]u(n)
s(n)= ay1 (1) +by2(n) (2.90)

Eq (2.89) and Eq (2.90) are equal. Therefore, the system is linear.

(c) The output depends on present input. Hence the system is causal.
() Given y(n) =x()u(n)
The output due to delayed input
y(n, ) = Tk(n-)]=x(n-k)u(n) (2.91)

The delayed output by ksamples is

y(n-k) =x(n-)u(n -) (2.92)
y(n, k) #y(n-4)
Hence the system is time-variant. ,
2.28 Signals and Systems

Practice Problem 2.4

Check whether the following systems are
(a) staticldynamic.
(b) lincar/non-lincar.
(c) causal/non-causal.
(d) time variant/time-invariant.
()y(n) =x(n+1) -x(n- 1)
(i) d'y),,dyt)dy)
+4 +2(1) =xt +5)
d13 di2 dt
(ii)y) = Evr()) Ans-) Dynamic, Linear, Non-causal, Time-invariant
(iv) y() = 10x{t+5 (ii) Dynamic, Nonlinear, Non-causal, Time-invariant
(iii) Dynamic, Linear, Non-causal, Time-variant
(v) (n) =k(n)| (iv) Static, Non-linear, Causal, Time-invariant
(vi) y(n) = d'x(n) (v) Static, Non-linear, Causal, Time-invariant
(vi) Static, Linear, Causal, Time-variant

Solved Problem 2.7 Check whether the following systems are

(a) Static/dynamic.
(b) linear/non-linear.
(c) Time-invariant or time-variant.
(i) A full wave rectifier whose output is the modulus of the input

y() = la()|
(ü)Amodulator whose carrier frequency is le. giving an output

y() =x() cos l,t

Solution:

() (a) The output at any instant depends on the input at that instant. Ther
fore the system is static,
(b) Given y) = k(0)|
For an input x()
(29)
n)=Th0)=n()|
For an input x()
(294)
nl) =Ta()) =k2(0)|
 Introduction to Systems 2.29

The output due to weighted sum of inputs is

(2.95)
vst) = Tlax () +brz(t)] = lax1(t) +bxlt)|
The weighted sum of outputs is
(2.96)
ay1 (t) +byzt) =alx1(1)| + b)z()|
ys(t) # ay1 () +byz(?)
Therefore, the system is non-linear.
(c) For an input x(1)
yl) = k()|
If the input is delayed by T seconds, the output is
y(t, T) = x(t-T)| (2.97)

If theoutput is delayed by T seconds we get

yt-T) =lk(t-T)| (2.98)
y(t, T) =y(t-T)
Hence the system is time-invariant.
(i) (a) Given y(:) = x() cos2,f
The output at any instant depends on the input at that instant. Hence
the system is static.
(b) Given y() =Tr()]=x() cos 2t
For an input xj()
yn() =T()]=() cosS2,t (2.99)

For an input x>()

y2) =Tn)] =n() cos 2,! (2.100)

The output due to weighted sum of inputs is

s() =Tlax() +bxn()]l (2.101)
= laxi() +brz()] cos 2et
= axn(t) cos ltt bxn() cos 2,t
The weighted sum of outputs is
ay1(t) +byz() =ax() cos 24 +bxg(() cosQ,t (2.102)
s() = ayn(1) + byal)
Eq (2.101) and Eq (2.102) are equal. Hence the system is linear.
2.30 Signals and Systems

(c) ) = Tla()] = x() cos 2

input is
The output due to delayed
Tx(-T)] =x(t-T)cos S2 (2.16,
ylr,T) =

The delayed output is

yl-T) =xl-T) cos2.(t-T) (2.104
vlt, T) #ylt-T)

time-variant.
Hence the system is

Practice Problem 2.5

Find whether the following systems are

a) Staticdynamic.
b) Linear/non-linear.
c) Time-imvariant or time-variant.

) y) =art) +btx(t -2) Ans: Dynamic, Linear, Time-invariant

(ii) y)=Ar) +B Ans: Static, Non-linear, Time-invariant
1
(ü) yo) = 2x(n) + Ans: Dynamic, Non linear, Time-invariant
x(7-1)
1
(Gv) yl(n) =P)+P-1) Ans: Dynamic, Non linear, Time-invariant

Revlew Questions

I. State superposition principle.

2. How do you find whether the system is
linear or nonlinear?
3. How do you prove that the system is AU Nov, 2008
4. Explain the following
time-invariant?
properties of the system.
(a) Stability
(c) Causality
(b) Linearity
() Time-invariance