Properties - of - Convolution
Properties - of - Convolution
CONVOLUTION
X(n) Y(n)
H(n)
∞
y(n) = x(n) * h(n) ≡ Σ x(k) h(n – k)
k=-∞
or
∞
y(n) = h(n) * x(n) ≡ Σ h(k) x(n – k)
k=-∞
“ * “ asterisk is used to simplify the notation for
convolution
Commutative Law
x(n) * h(n) = h(n) * x(n)
Graphical Representation:
X(n) Y(n) H(n) Y(n)
H(n) H(n)
Assosciative Law
[x(n) * h1(n)] * h2(n) = x(n) * [h1(n)
* h2(n)]
Graphical Representation:
x(n) y1(n) y(n)
H1(n) H2(n)
x(n) y(n)
H2(n)
x(n) y(n)
H(n) =h1(n) +
h2(n)
Causal Linear Time Invariant Systems
Given the convolution formula:
∞
y(no) = Σ h(k) x(no – k)
k=-∞
divide into two (2) sets of terms:
∞ -1
y(no) = Σ x(k) h(no – k) + Σ h(k) x(no – k)
k=0 k=-∞
= [h(0)x(no) + h(1)x(no – 1) + h(2)x(no – 2) + …]
+ [h(-1)x(no + 1) + (-2)x(no + 2) + …]
∞
y(n) = h(n) * x(n) ≡ Σ h(k) x(n – k)
k=-∞
and get the absolute value for both sides
∞ ∞
|y(n)| = | Σ h(k) x(n – k) | = Σ |h(k)| |x(n – k)|
k=-∞ k=-∞
substituting this equation to equation 1 form:
∞
|y(n)| ≤ Mx Σ h(k) x
k=-∞
A linear time invariant system is stable if its impulse response is absolutely
summable .
Systems with Finite-duration and Infinite
Duration Impulse Response
2 types of Linear Time invariant Systems
We need the value y(no – 1) called the initial condition and the
input samples x(n) for n ≥ no. the initial condition contains all the
information needed to determine the response of the system for n
≥ no to the input signal x(n), independent of the past occurrences.
Z-1
The presence of the feedback loop in the recursive system contains a
delay element and is crucial for the realization of the system since the
absence would force the system to compute y(n) in terms of y(n) which
is not possible for discrete time systems.
In summary, the output of a recursive system should be computed in
order (y(0), y(1), y(2), …) whereas non-recursive system, the output
can be computed in any order [y(2000), y(15), y(3), y(100),…].
Realization of recursive systems
A. Cumulative average of a signal
n
y(n) = [1/(n + 1)] Σ x(k) n = 0, 1, …
k + +
1
n
n+
(n + 1) y(n) = Σ x(k)
1
k=0 Z-1
+
Then: = ny(n – 1) + x(n)
y(n) = [n/(n + 1)] y(n – 1) + [1/(n + 1)] x(n)
N M
Σ ak y(n – k) = Σ bk x(n
If the constants are time dependent,
– k) then time variant. Therefore, the system
defined by a LTI constant coefficient difference equation is linear and time
invariant. k=0 k=0
By just looking at the mathematical equation of the input and output
relationship the system is causal.
Last stability. BIBO if and only if for every bounded input and every bounded
initial condition, the total system response is bounded output.
Solution to Linear Constant- Coefficient
Difference Equation
The objective is to determine an explicit
expression for the output y(n). 2 methods, the Direct
Method and the indirect method (z transform).
N M
Σ ak y(n – k) = Σ bk x(n – k)
a0 = 1
k=0 k=0
depends on the form of the input x(n).
The total solution of the difference equation
X(n y(n)
) + +
b0 v(n)
z-1 z-1
b1 -a1
The realization uses separate delays (memory) for both the input and the output
signal samples and is called the direct form I structure. Note – the system can be
viewed as 2 linear time invariant systems in cascade. The first is non-recursive and
the second recursive.
v(n) = b)x(n) + b1x(n – 1)
y(n) = -a1y(n – 1) + v(n)
Interchanging the order of a cascaded LTI system will not change the overall response.
w(n) = -a1w(n – 1) + x(n)
y(n) = b0w(n) + b1w(n – 1)
2 difference equation are equivalent to the single difference
equation
X(n Y(n)
) + +
W(n)
b0
Z -1
Z -1
-a1 b1
There are 2 delay elements containing the same input w(n) which is an
auxiliary quantity and these 2 can be merge into one delay element. The
realization is more efficient in terms of memory requirements and is called
direct form II structure used extensively in practical applications
+ +
X(n W(n) b0 Y(n)
) Z-1
-a1 b1
Correlation of discrete time signals
Mathematically:
y(n) = ax(n – D) + w(n)