The Dirac Delta Function and Convolution
The Dirac Delta Function and Convolution
The Dirac Delta Function and Convolution
dX(x) d(t)
1/T
1
1/T
2
1/T
3
1/T 4
t t
0 T T T T 0
1 2 3 4
a) Unit pulses of different extents b) The impulse function
Figure 1 shows a unit pulse function δT (t), that is a brief rectangular pulse function of duration
T , defined to have a constant amplitude 1/T over its extent, so that the area T × 1/T under the
pulse is unity:
0 for t ≤ 0
δT (t) = 1/T 0<t≤T (3)
0 for t > 0.
The Dirac delta function (also known as the impulse function) can be defined as the limiting form
of the unit pulse δT (t) as the duration T approaches zero. As the duration T of δT (t) decreases,
the amplitude of the pulse increases to maintain the requirement of unit area under the function,
and
δ(t) = lim δT (t). (4)
T →0
The impulse is therefore defined to exist only at time t = 0, and although its value is strictly
undefined at that time, it must tend toward infinity so as to maintain the property of unit area in
the limit. The strength of a scaled impulse Kδ(t) is defined by its area K.
The limiting form of many other functions may be used to approximate the impulse. Common
functions include triangular, gaussian, and sinc (sin(x)/x) functions.
1
The impulse function is used extensively in the study of linear systems, both spatial and tem-
poral. Although true impulse functions are not found in nature, they are approximated by short
duration, high amplitude phenomena such as a hammer impact on a structure, or a lightning strike
on a radio antenna. As we will see below, the response of a causal linear system to an impulse
defines its response to all inputs.
An impulse occurring at t = a is δ(t − a).
This is easily seen by noting that δ(t − a) is zero except at t = a, and for its infinitesimal duration
f (t) may be considered a constant and taken outside the integral, so that
∞ ∞
f (t)δ(t − a)dt = f (a) δ(t − a)dt = f (a) (6)
−∞ −∞
2 Convolution
Consider a linear continuous-time system with input u(t), and response y(t), as shown in Fig. 2.
We assume that the system is initially at rest, that is all initial conditions are zero at time t = 0,
and examine the time-domain forced response y(t) to a continuous input waveform u(t).
In Fig. 3 an arbitrary continuous input function u(t) has been approximated by a staircase
function ũT (t) ≈ u(t), consisting of a series of piecewise constant sections each of an arbitrary
fixed duration, T , where
for all n. It can be seen from Fig. 3 that as the interval T is reduced, the approximation becomes
more exact, and in the limit
u(t) = lim ũT (t).
T →0
The staircase approximation ũT (t) may be considered to be a sum of non-overlapping delayed pulses
pn (t), each with duration T but with a different amplitude u(nT ):
∞
ũT (t) = pn (t) (8)
n=−∞
2
u(t) u(t)
u~T(t)
System Input
System Input
u(t)
u(3T)
u~T(t)
0 t 0 t
0 T 2T 3T .... 0 T 2T 3T ....
time time
dT( t - n T ) yT (t)
1/T
dT( t - n T ) system yT (t)
0 t 0
0 nT (n+1)T t
0 nT (n+1)T
Total Response
yi (t) ~
y (t)
T
0 t 0 t
0 T 2T 3T .... time 0 T 2T 3T .... Time
3
where
u(nT ) nT ≤ t < (n + 1)T
pn (t) = (9)
0 otherwise
Each component pulse pn (t) may be written in terms of a delayed unit pulse δT (t) defined in Sec.
1, that is:
pn (t) = u(nT )δT (t − nT )T (10)
so that Eq. (8) may be written:
∞
ũT (t) = u(nT )δT (t − nT )T. (11)
n=−∞
We now assume that the system response to δT (t) is a known function and is designated hT (t)
as shown in Fig. 4. Then if the system is linear and time-invariant, the response to a delayed unit
pulse, occurring at time nT , is simply a delayed version of the pulse response:
yn (t) = hT (t − nT ). (12)
The principle of superposition allows the total system response to ũT (t) to be written as the sum
of the responses to all of the component weighted pulses in Eq. (11):
∞
ỹT (t) = u(nT )hT (t − nT )T (13)
n=−∞
as shown in Fig. 5. For physical systems the pulse response hT (t) is zero for time t < 0, and future
components of the input do not contribute to the sum, so that the upper limit of the summation
may be rewritten:
N
ỹT (t) = u(nT )hT (t − nT )T for N T ≤ t < (N + 1)T. (14)
n=−∞
Equation (14) expresses the system response to the staircase approximation of the input in terms
of the system pulse response hT (t). If we now let the pulse width T become very small, and write
nT = τ , T = dτ , and note that limT →0 δT (t) = δ(t), the summation becomes an integral:
N
y(t) = lim u(nT )hT (t − nT )T (15)
T →0
n=−∞
t
= u(τ )h(t − τ )dτ (16)
−∞
Equation (16) is an important integral in the study of linear systems and is known as the convolution
or superposition integral. It states that the system is entirely characterized by its response to an
impulse function δ(t), in the sense that the forced response to any arbitrary input u(t) may be
computed from knowledge of the impulse response alone. The convolution operation is often written
using the symbol ⊗: t
y(t) = u(t) ⊗ h(t) = u(τ )h(t − τ )dτ. (18)
−∞
4
System impulse response
h(t) h(-t)
time reversal
t t
0
time
shifting
System input
u(t)
h(t -t)
1
t t
0 0 t1
multiplication
u(t)h(t-t)
t
0 t
1
System response
y(t)
0 t1 t
5
Equation (18) is in the form of a linear operator, in that it transforms, or maps, an input function
to an output function through a linear operation. It is a direct computational form of the system
transfer operator H {u(t)}, that is:
The form of the integral in Eq. (16) is difficult to interpret because it contains the term h(t − τ ) in
which the variable of integration has been negated. The steps implicitly involved in computing the
convolution integral may be demonstrated graphically as in Fig. 6, in which the impulse response
h(τ ) is reflected about the origin to create h(−τ ), and then shifted to the right by t to form
h(t − τ ). The product u(t)h(t − τ ) is then evaluated and integrated to find the response. This
graphical representation is useful for defining the limits necessary in the integration. For example,
since for a physical system the impulse response h(t) is zero for all t < 0, the reflected and shifted
impulse response h(t − τ ) will be zero for all time τ > t. The upper limit in the integral is then
at most t. If in addition the input u(t) is time limited, that is u(t) ≡ 0 for t < t1 and t > t2 , the
limits are:
t
u(τ )h(t − τ )dτ for t < t2
yf (t) = t1t2 (19)
u(τ )h(t − τ )dτ for t ≥ t2
t1
Example
A mass element, shown in Fig. 7 at rest on a viscous plane, is subjected to a very short
unit impulsive force of duration 0.001 seconds and magnitude 1000 newtons, and is
observed to respond with a velocity vm (t) = e−3t . Find the response of the same mass
F(t) m 1
vm(t)
t 0 t
0 time 0 1 2
time (sec)
6
Parallel systems:
t
= e−3t τ e3τ dτ (22)
0
where the limits have been chosen because the system is causal, and the input is iden-
tically zero for all t < 0. Integration by parts gives the solution
1 1 1
v(t) = t − + e−3t . (23)
3 9 9
The associative property may be interpreted as an expression for the response on two systems in
cascade or series, and indicates that the impulse response of two systems is h1 (t) ⊗ h2 (t), as shown
in Fig. 8. Similarly the distributive property may be interpreted as the impulse response of two
systems connected in parallel, and that the equivalent impulse response is h1 (t) + h2 (t).