Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd
Upload
154 views

ADSP

This document discusses discrete convolution and properties of the convolution integral in signal processing. It begins with an example of numerical convolution using MATLAB by defining input signals x[n] and h[n], and using the conv function. It then covers the convolution representation of linear time-invariant systems and the convolution integral. Properties of the convolution integral discussed include associativity, commutativity, distributivity, shift property, differentiation property, and integration property. The document also discusses representing a continuous-time linear time-invariant system using its unit-step response.

Uploaded by

Moez Ul Hassan
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online on Scribd
154 views

ADSP

This document discusses discrete convolution and properties of the convolution integral in signal processing. It begins with an example of numerical convolution using MATLAB by defining input signals x[n] and h[n], and using the conv function. It then covers the convolution representation of linear time-invariant systems and the convolution integral. Properties of the convolution integral discussed include associativity, commutativity, distributivity, shift property, differentiation property, and integration property. The document also discusses representing a continuous-time linear time-invariant system using its unit-step response.

Uploaded by

Moez Ul Hassan
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 41

SP503

Advanced Digital Signal Processing

Week # 08: 11th November, , 2013

Engr.AHMEDALI

DiscreteConvolutionExample
Problem#01:

DiscreteConvolutionExample
Solution

DiscreteConvolutionExample
Solution

DiscreteConvolutionExample
Solution

DiscreteConvolutionExample
Solution

DiscreteConvolutionExample
Solution

DiscreteConvolutionExample
Solution

DiscreteConvolutionExample
Solution

DiscreteConvolutionExample
Solution

DiscreteConvolutionExample
Solution

DiscreteConvolutionExample
Solution

DiscreteConvolutionExample
FinalOutput

NumericalConvolutionUsingMATLAB
Definex[n]andh[n]
xx = [5 7 -9 10]; hh = [1 2 3]; Mayalsobeusefultousethezeros andones functions(especiallyformatchingupsample values)

Usetheconv function
yy = conv(xx,hh);

DiscreteConvolutionExample

DiscreteConvolutionExample
Solution(a)

DiscreteConvolutionExample

DiscreteConvolutionExample

DiscreteConvolutionExample

DiscreteConvolutionExample

DiscreteConvolutionExample

CTUnitImpulse p Response p ConsidertheCTSISOsystem:

x(t )

System

y (t )

Iftheinputsignalis x(t ) = (t )andthe systemhasnoenergyat t = 0 ,theoutput y (t ) = h(t )iscalledtheimpulseresponseof thesystem

(t )

System y

h(t )

Exploiting p gTimeInvariance Letx[n]beanarbitraryinputsignalwith x(t ) = 0, for t < 0 Usingthesiftingpropertyof (t ),wemay write

x(t ) =

x( ) (t )d ,
System

t0

Exploitingtimeinvariance invariance,itis

(t )

h(t )

Exploiting p g Time-Invariance

Exploiting p g Linearity y Exploitinglinearity linearity, , itis

y (t ) =

x( )h(t )doesnotcontain Iftheintegrand g animpulselocatedat = 0,thelowerlimitof g canbetakentobe0,i.e., theintegral

x( )h(t )d ,

t0

y (t ) = x( )h(t )d , t 0
0

The Convolution Integral Thisparticularintegrationiscalledthe convolution l i integral i l

y (t ) = x( )h(t )d , t 0 0 144 244 3 x(t ) h(t ) Equation iscalledthe y (t ) = x(t ) h(t )


convolutionrepresentationofthesystem Remark:aCTLTIsystemiscompletely d described ib dby b its it impulse i l responseh(t)

Block Diagram Representation of CT LTI Systems Sincetheimpulseresponseh(t)providesthe completedescriptionofaCTLTIsystem,we write

x(t )

h(t )

y (t )

Example: Analytical Computation of the Convolution Integral Supposethatwherep(t)is x( t) = h (t ) = p (t ), figure therectangular pulse depicted in

p (t )

Example Contd Inordertocomputetheconvolution integral

y (t ) = x( )h(t )d , t 0
0

wehavetoconsiderfourcases:

Example Contd Case1: t 0

h(t )
t T

x( )

0
y (t ) = 0

Example Contd Case2: 0 t T

h(t )

x( )

t T

0 t
t

y (t ) = d = t
0

Example Contd Case3: 0 t T T

T t 2T
h(t )

x( )

0 t T
T

y (t ) =

t T

d = T (t T ) = 2T t

Example Contd Case4:

T t T x( )

2T t
h(t ) T t T

0
y (t ) = 0

Example Cont Contd d

y (t ) = x(t ) h(t )

2T

Properties p of the Convolution Integral g Associativity

x(t ) (v(t ) w(t )) = ( x(t ) v(t )) w(t )


Commutativity

x(t ) v(t ) = v(t ) x(t )


Distributivity yw.r.t.addition

x(t ) (v(t ) + w(t )) = x(t ) v(t ) + x(t ) w(t )

Properties of the Convolution Integral g - Contd Shiftproperty: define then

w(t q ) = xq (t ) v(t ) = x(t ) vq (t )


Convolutionwiththeunitimpulse

xq (t ) = x(t q ) v (t ) = v(t q ) q w(t ) = x(t ) v(t )

x(t ) (t ) = x(t )
Convolutionwiththeshiftedunitimpulse

x(t ) q (t ) = x(t q )

Properties of the Convolution Integral - Cont Contd d Derivativeproperty:ifthesignalx(t)is differentiable,thenitis

d dx(t ) v(t ) [ x(t ) v(t )] = dt dt


Ifb both thx(t)and dv(t)aredifferentiable, diff ti bl th then itisalso

d dx(t ) dv(t ) x(t ) v(t ) ] = 2[ dt dt dt

Properties of the Convolution Integral - Cont Contd d Integration I t ti property: t define d fi t ( 1)) x (t ) x( )d t v ( 1) (t ) v( )d then

( x v)

( 1)

(t ) = x

( 1)

(t ) v(t ) = x(t ) v

( 1)

(t )

Representation of a CT LTI System in Terms of the Unit-Step Response Letg(t)betheresponseofasystemwith impulseresponseh(t)whenx(t ) = u (t ) with noinitialenergyattime ie t = 0,i.e.,

u (t )
Therefore,itis

h(t )

g (t )

g (t ) = h(t ) u (t )

Representation of a CT LTI System in Terms of the Unit-Step Response Contd Differentiatingbothsides

dg (t ) dh(t ) du (t ) = u (t ) = h(t ) dt dt dt
Recallingthat

du (t ) = (t ) dt
itis

and

h(t ) = h(t ) (t ) g (t ) = h( )d
0
t

dg (t ) = h(t ) dt

or

QUIZ#04
QUESTION: [05marks]

41

You might also like