3150912: SIGNALS & SYSTEMS

Q.1 List various type of systems and define them giving example.

Various types of systems are listed as follows

1. Linear & Non-linear systems
2. Time invariant & Time variant systems
3. Causal & Non-causal systems
4. Stable & Unstable systems
5. Static & Dynamic systems
6. Invertible & Non-invertible systems
The systems are defined as follows.
1. Linear & Non-linear systems
• System that satisfies two properties: 1) Additivity and 2) Homogeneity is called Linear system.

x(t) System T y(t) = T{x(t)}

If T{ax1(t)+bx2(t)} = aT{x1(t)} + bT{x2(t)} = ay1(t)+by2(t) then system is Linear.

Example of linear system : CT: y(t) = x(t) DT: y(n) = x(n) + x(n-1)

• System that does not satisfies these properties is called Non-linear system.

Example of Non-linear system : CT: y(t) = x2(t) DT: y(n) = 2 x(n) + 3

2. Time Invariant & Time variant systems

• If the output of the system does not change with respect to time provided the input is constant is
called Time invariant systems.
x(t) System T y(t) = T{x(t)}

If T{x(t-T)} = y(t-T) then system is Time invariant system.

Example of Time invariant system : CT: y(t) = x(t) DT: y(n) = x(n) + x(n-1)

• System that does not satisfies above property then it is called Time variant system.

Example of Time variant system : CT: y(t) = tx(t) DT: y(n) = x(2n)

3. Causal & Non-causal systems

• If the present value of system output depends only on the preset and past values of system input then
the system is called Causal system.

Example of causal system : CT: y(t) = x(t) DT: y(n) = x(n) + x(n-1)

• If the present value of system output also depends on the future values of input then the system is
called Non-causal system.

Example of Non-causal system : CT: y(t) = x(t+1) DT: y(n) = x(-n)

Department of EE and EC
Darshan Institute of Engineering & Technology, Rajkot
3150912: SIGNALS & SYSTEMS
4. Static (Memoryless) & Dynamic system
• If present value of system output depends only on the preset value of system input then the system is
called Static system. Such system does not require memory and there is no need of past of future
values, hence they are also called as Memoryless systems.

Example of causal system : CT: y(t) = x(t) DT: y(n) = x(n) + x(n-1)

• If the present value of system output depends on past and/or future values of input then the system is
called Dynamic system.

Example of Dynamic system : CT: y(t) = x(t-1) DT: y(n) = x(-n)

• Static systems are always causal but all causal systems are not static.

5. Stable & Unstable system

• For every bounded input if the output of the system is also bounded then the system is called Stable
system.(BIBO stability criteria)

Example of stable system : CT: y(t) = 𝑠𝑖𝑛𝜔𝑡x(t) DT: y(n) = x(n) + x(n-1)

• If the above condition is not satisfied then the system is called Unstable system.

Example of Dynamic system : CT: y(t) = t .x(t-1) DT: y(n) = (n+1) x(n)

6. Invertible & Non-invertible system

• If it is possible to obtain input signal from the output of the system then the system is called Invertible
system. This is shown in figure below.
Inverse x(t)
x(t) System T y(t) System T-1

Example of Invertible system : CT: y(t) = 𝑥(𝑡)dt DT: y(n) = 2x(n)

• If the above condition is not satisfied then the system is called Non-invertible system.

Example of Dynamic system : CT: y(t) = x2 (t) DT: y(n) = nx(n)

Q.2 Define impulse response and state its importance.

• Impulse response h(t) or h(n) for an LTI system is defined as output of the system when the input is
impulse signal i.e. δ(t) or δ(n).
δ(t) System T y(t) = T{δ(t)} = h(t)

• Impulse response h(t) or h(n) completely describe an LTI system. This means that, if h(t) or h(n) of a
system is known then output of the system for any arbitrary input can be calculated using convolution
relation.
System y(t) = T{x(t)} = x(t)*h(t)
x(t)
h(t)
Department of EE and EC
Darshan Institute of Engineering & Technology, Rajkot
3150912: SIGNALS & SYSTEMS
Q.3 Define convolution in CT & DT. List its property and prove that convolution is commutative.
• Convolution relations in CT and DT domain are defined as

Discrete Time: x ( n) * h( n)   x ( k ) h( n  k )
k  

Continuous Time: x(t ) * h(t )   x( )h(t   )d


• Properties of convolution

1. Commutative property:
x ( n) * h( n)  h( n) * x ( n)
x(t ) * h(t )  h(t ) * x(t )
2. Associative property:
{x1 (n) * x2 (n)}* x3 (n)  x1 (n) *{x2 (n)}* x3 (n)}
{x1 (t ) * x2 (t )}* x3 (t )  x1 (t ) *{x2 (t )}* x3 (t )}
3. Distributive property:
x1 (n) *{x2 (n)  x3 (n)}  x1 (n) * x2 (n)  x1 (n) * x3 (n)
x1 (t ) *{x2 (t )  x3 (t )}  x1 (t ) * x2 (t )  x1 (t ) * x3 (t )

4. Convolution with impulse:

x ( n) *  ( n)  x ( n)
x(n) *  (n  n0 )  x(n  n0 )
x(t ) *  (t )  x(t )
x(t ) *  (t  T0 )  x(t  T0 )
• Prove convolution is commutative

x ( n) * h( n)   x ( k ) h( n  k )
k  
Let, 𝑛 − 𝑘 = 𝑚 ⇒ 𝑘 = 𝑛 − 𝑚
𝑘 → −∞ ⇒ 𝑚 → ∞ 𝑎𝑛𝑑 𝑘 → ∞ ⇒ 𝑚 → −∞

x ( n) * h( n)   x(n  m)h(m)
m 

  h(m) x(n  m)
m  
(addition and multiplication are commutative)

 h(n) * x(n)

Hence proved that convolution is commutative. Similarly it can be proved for CT domain

Department of EE and EC
Darshan Institute of Engineering & Technology, Rajkot
 3150912: SIGNALS & SYSTEMS
Q.4 Derive the condition on impulse response for LTI system to be Causal System

• Consider Discrete time LTI system as follows

x(n) System y(n) = T{x(n)}

h(n)

• Using convolution relation output of the system can be expressed in term of impulse response as

y ( n)  x ( n) * h( n)   x ( k ) h( n  k )
k  

  h( k ) x ( n  k )
k  
(convolution is commutative)

y(n)  {.... h(2) x(n  2)  h(1) x(n  1)}  {h(0) x(n)   h(1) x(n  1)   h(2) x(n  2)  ...}

• Causal system is defined as system whose present value of output depends only on the past and
present values of input.
• Here in above equation, terms in the first bracket indicates dependency on future value of input and
• terms in second bracket indicates dependency on present and past values of input.
• Thus in this equation if h(n) =0 for n<0 then y(n) depends only on the second bracket terms which
makes system causal.
• Therefore, condition on impulse response for LTI system to be causal is

h(n) =0 for n<0

• Similarly for CT system to be causal condition will be

h(t) =0 for t<0

Q.5 Derive the condition on impulse response for LTI system to be Stable System

• Consider Discrete time LTI system as follows

x(n) System y(n) = T{x(n)}

h(n)

• Using convolution relation output of the system can be expressed in term of impulse response as

y ( n)  x ( n) * h( n)   x ( k ) h( n  k )
k  

  h( k ) x ( n  k )
k  
(convolution is commutative)

• System is stable if for every bounded input output of the system is also bounded (BIBO Criteria).

• Let the input x(n) be bounded i.e.

𝑥 𝑛 ≤ 𝑀𝑥 ≤ ∞

Department of EE and EC
Darshan Institute of Engineering & Technology, Rajkot
3150912: SIGNALS & SYSTEMS
• To find condition for output y(n) to be bounded consider -
∞

|𝑦 𝑛 | = ℎ 𝑘 𝑥 𝑛−𝑘
𝑘=−∞
∞

|𝑦 𝑛 | ≤ |ℎ 𝑘 𝑥 𝑛−𝑘 | { 𝑥1 + 𝑥2 ≤ |𝑥1| + |𝑥2| }

𝑘=−∞
∞

|𝑦 𝑛 | ≤ ℎ 𝑘 |𝑥 𝑛 − 𝑘 | { 𝑥1 ∙ 𝑥2 = |𝑥1| ∙ |𝑥2| }
𝑘=−∞
• Input is bounded and has maximum value of Mx
∞

|𝑦 𝑛 | ≤ 𝑀𝑥 ℎ 𝑘
𝑘=−∞ ∞

• From above equation output will be bounded if ℎ 𝑘 is finite.

𝑘=−∞
• Therefore, condition on impulse response for LTI system to be stable is

ℎ 𝑛 <∞
𝑛=−∞

• Similarly for CT system to be causal condition will be

∞
ℎ 𝑡 𝑑𝑡 < ∞
−∞

Q.6 Derive the condition on impulse response for LTI system to be Invertible System

• System is said to be invertible if it is possible to obtain input signal from the output of the given
system by using its inverse system.

Inverse
System
x(n) System y(n)=x(n)
h1(n) y1(n)
h2(n)
• Here
Output of System 1 - 𝑦1 𝑛 = 𝑥 𝑛 ∗ ℎ1 (n)
Output of System 2 - 𝑦 𝑛 = 𝑦1 𝑛 ∗ ℎ2 𝑛 = {𝑥 𝑛 ∗ ℎ1 𝑛 } ∗ ℎ2 𝑛
𝑦 𝑛 = 𝑥 𝑛 ∗ {ℎ1 𝑛 ∗ ℎ2 𝑛 } (convolution satisfies Associative Law)
• Now as per convolution property
x ( n) *  ( n)  x ( n)
• So in above equation y(n) = x(n) if and only if ℎ1 𝑛 ∗ ℎ2 𝑛 = δ(𝑛)
• Therefore, two systems from pair of invertible system if
ℎ1 𝑛 ∗ ℎ2 𝑛 = δ(𝑛)

Department of EE and EC
Darshan Institute of Engineering & Technology, Rajkot
3150912: SIGNALS & SYSTEMS
Q.7 State and prove Sampling Theorem. OR What is aliasing? Explain in detail.
Sampling Theorem - Statement
“A continuous time signal x(t) can be properly represented and recovered from its discrete time sampled
representation if the sampling frequency 𝑓𝑠 ≥ 2𝑊 where W is the largest frequency of signal x(t).”
If this condition is not satisfied then signal of one frequency makes itself appear as signal of some another
frequency and hence becomes indistinguishable. This problem is called the problem of Aliasing.
Proof of Sampling Theorem
• Consider continuous time signal x(t) to be band-limited having highest frequency component of W Hz.
Let 𝑋(𝐹) represents frequency spectrum of signal x(t).

x(t) x(n)

s(t)

• Consider a train of impulse signal s(t) as shown in the figure where impulse signal occurs at every
integer multiple of Ts.
• The process of sampling can be considered as multiplication of x(t) with s(t) i.e.
discrete time sampled signal x(n) = x(t)·s(t)
1
Here 𝑇𝑠 is the sampling period and 𝑓𝑠 = is the sampling frequency.
𝑇𝑠
• As per property of Fourier Transform - F.T{ x(t)·s(t) } = X(F) * S(F)
• Now Fourier Transform S(F) of s(t) will be as shown in the figure and it
can be expressed as. ∞

𝑆 𝐹 = 𝛿(𝐹 − 𝑘𝑓𝑠 )
𝑘=−∞
∞ ∞ ∞

𝑋 𝐹 ∗𝑆 𝐹 =𝑋 𝐹 ∗ 𝛿 𝐹 − 𝑘𝑓𝑠 = 𝑋 𝐹 ∗ 𝛿 𝐹 − 𝑘𝑓𝑠 = 𝑋 𝐹 − 𝑘𝑓𝑠

𝑘=−∞ 𝑘=−∞ 𝑘=−∞

• Now for different values of W and 𝑓𝑠 frequency spectrum of discrete time sampled signal will be as
follows. 𝑓 > 2𝑊
𝑠

Signal recoverable

𝑓𝑠 = 2𝑊

Signal recoverable

Aliasing effect 𝑓𝑠 < 2𝑊

Signal not recoverable

• From above it can be observed that from the spectrum of sampled signal, original signal can be
recovered in case 1 & 2 using ideal low pass filter. In case 3 due to aliasing effect the original signal
can not be recovered.
• Hence as per Sampling theorem 𝑓𝑠 ≥ 2𝑊
Department of EE and EC
Darshan Institute of Engineering & Technology, Rajkot