Lecture 9: Convolution Evaluation: Continuous time system In this lecture we shall solve the convolution integral using different
methods. 1 Methods of evaluating the convolution integral:
Given the system impulse response h(t), and the input x(t), the system output y(t), is given by the convolution integral:
y (t ) =
x( )h(t )d
(2)
Some of the different methods of evaluating the convolution integral are: Graphical representation, Mathematical equation, Laplace-transforms, Fourier Transform, Differential equation, Block diagram representation, and finally by going to the digital domain. We shall now see each of these methods. 1.1 Evaluation using graphical representation A method of computing the convolution integral is through the direct computation of each value of the output y[t]. This method is based on evaluation of the convolution integral for a single value of t, and varying t over a number of possible values, to enable plot of a smooth curve.
y (t ) =
Step 1: Step 2: Step 3: Sketch x( )
Sketch h( ) Compute y(0) using:
x( )h(t )d
y (0) =
x( )h( )d
which is the integral of the product of the two signals x( ) & h( )
Step 4: Step 5:
Sketch h(t0 ) , which is right shift of h( ) by t0 . Compute y( t0 ) using:
y (t0 ) =
x( )h(t0 )d
which is the integral of the product of the two signals x( ) & h(t0 )
Step 6: Step 7: Step 8:
Repeat steps 4 & 5, for a few different values of t0 , so that a smooth curve for y(t), for positive t, can be sketched. Sketch h(t0 ) , which is left shift of h( ) by t0 . Compute y(- t0 ) using:
y ( t0 ) =
x( )h(t0 )d
which is the integral of the product of the two signals x( ) & h(t0 ) .
Step 9: Step 10: 1.2
Repeat step 7 & 8, for a few different values of t0 , so that a smooth curve for y(t), for negative t, can be sketched. Sketch the computed results of all the above steps to obtain the system output y(t)
Evaluation from the convolution integral
While small length, finite duration sequences can be convolved using graphical representation, when the signals to be convolved are of infinite length, the convolution is easier performed by direct use of the convolution integral of equation (2). It may be recalled that:
u ( ) =
0 1 0 1 0 1
0 1
for < 0 for 0
for (t ) < 0 for (t ) 0 for ( ) < t for ( ) t for > t for t
u (t ) = = =
Example: A system has impulse response h(t ) = exp(0.8t )u (t ) . Obtain the unit step response. Solution:
y (t ) = = = =
x( )h(t )d
{u ( )}{exp(0.8(t ))u (t )} d
{exp(0.8(t ))u (t )} d {exp(0.8(t ))} d
t 0
0 t
= exp(0.8t ) {exp(0.8 )} d =
[1 exp(0.8t )]
(0.8)
1.3
Evaluation using Laplace-transforms
Another method of computing the convolution of two sequences is through use of Laplace -transforms. This approach converts convolution to multiplication in the transformed domain.
1.4 Evaluation from Fourier transform (FT)
It is possible to compute the convolution of two sequences by transforming them to the frequency domain through application of the Fourier Transform. This approach also converts the convolution operator to multiplication.
1.5 Evaluation from block diagram representation
While small length, finite duration sequences can be convolved by any of the above three methods, when the sequences to be convolved are of infinite length, the convolution is easier performed by direct use of the convolution sum..
1.6 Evaluation by recourse to digital domain
It is possible to compute the convolution of two analog signals by using the discreteconvolution sum. The advantage being that it is relatively easy to write a computer program for discrete time convolution rather than the analog convolution integral. Hence each of the two signals may be sampled (in accordance to the sampling theorem), convolved in the digital domain, and then low pass filtering of the digital convolved sequence to obtain the analog output, which happens to be the convolution of the two given sequences.
A comment on the convolution operator
As discussed, the convolution operator, is a mathematical tool, that helps in computing the output of the digital system, given a specific input and the system impulse response. It is to be emphasized that input and output of a physical system are always x(t) and y(t), and the output depends on the impulse response of the system, all of them being in the TIME DOMAIN. Hence, physical quantities of any system exist only in the time domain.
