Digital Signals
For floating point DS processor, the amplitudes can be floating points.
CEN352, Dr. Ghulam Muhammad King Saud University
Common Digital Sequences
Unit-impulse sequence:
Unit-step sequence:
CEN352, Dr. Ghulam Muhammad King Saud University
Shifted Sequences
Shifted unit-impulse Shifted unit-step
Right shift by two samples
Left shift by two samples
CEN352, Dr. Ghulam Muhammad King Saud University
Sinusoidal and Exponential Sequences
Example
Sinusoidal
Example
CEN352, Dr. Ghulam Muhammad King Saud University
Exponential
Example 1
Solution:
CEN352, Dr. Ghulam Muhammad King Saud University
Generation of Digital Signals
Let, sampling interval, t T x(n): digital signal x(t): analog signal
Also
Example 2
Convert analog signal x(t) into digital signal x(n), when sampling period is 125 microsecond, also plot sample values.
Solution:
CEN352, Dr. Ghulam Muhammad King Saud University
Example 2 (contd.)
The first five sample values:
Plot of the digital sequence:
CEN352, Dr. Ghulam Muhammad King Saud University
Linear System
System: A system that produces an output signal in response to an input signal.
Continuous system & discrete system.
Time, t
Sample number, n
CEN352, Dr. Ghulam Muhammad King Saud University
Linear Systems: Property
1.
2. 3.
Homogeneity
Additivity Shift invariance
Must for all linear systems
Must for DSP linear systems
Homogeneity: (deals with amplitude) If x[n] y[n], then kx[n] ky[n] K is a constant v = iR v across R Linear System
p v2
i across R
v across R
Non Linear System
p in R
CEN352, Dr. Ghulam Muhammad King Saud University
Linear Systems: Property
Additivity
Homogeneity & Additivity
10
CEN352, Dr. Ghulam Muhammad King Saud University
Linear Systems: Property
Shift (time) Invariance
11
CEN352, Dr. Ghulam Muhammad King Saud University
Example 3
x(n) y(n)
Let a digital amplifier, If the inputs are: Outputs will be:
x1 (n) u(n)
y1 (n) 10u(n)
X 10
X 10
x2 (n) (n)
X 10
y2 (n) 10 (n)
If we apply combined input to the system: The output will be:
2 y1 (n) 2 10 x1 (n) 20u (n) 4 y2 (n) 4 10 x2 (n) 40 (n)
Individual outputs:
2 x1 (n) 2u(n)
12
Linear System
4 y2 (n) 40 (n)
X 10 CEN352, Dr. Ghulam Muhammad
King Saud University
2 y1 (n) 20u(n)
4 x2 (n) 4 (n)
X 10
Example 4
x(n)
y(n) x 2 (n)
System
x1 (n) u(n)
y1 (n) u 2 (n) u(n)
System
x2 (n) (n)
System
y2 (n) 2 (n) (n)
If the input is:
Then the output is:
Individual outputs:
4 y1 (n) 4 x1 2 (n) 4u (n) 2 y2 (n) 2 x2 (n) 2 (n)
2
Non Linear System
13
CEN352, Dr. Ghulam Muhammad King Saud University
Example 5 (a)
Given the linear system , find whether the system is time invariant or not.
Solution:
x1 (n)
System
y1 (n) 2 x1 (n 5)
Let the shifted input be:
Therefore system output: Shifting y1 (n) 2 x1 (n 5) by n0 samples leads to Equal
Time Invariant
14 CEN352, Dr. Ghulam Muhammad King Saud University
Example 5 (b)
Given the linear system y(n) 2 x(3n) , find whether the system is time invariant or not.
Solution:
x1 (n)
System
y1 (n) 2 x1 (3n)
Let the shifted input be: Therefore system output: Shifting
y1 (n) 2 x1 (3n)
NOT Equal
by n0 samples leads to
NOT Time Invariant
15 CEN352, Dr. Ghulam Muhammad King Saud University
Causality
Causal System:
Output y(n) at time n depends on current input x(n) at time n or previous inputs, such as x(n-1), x(n-2), etc.
Example:
Non Causal System:
Output y(n) at time n depends on future inputs, such as x(n+1), x(n+2), etc.
Example:
The non causal system cannot be realized in real time.
16 CEN352, Dr. Ghulam Muhammad King Saud University
Difference Equation
A causal, linear, and time invariant system can be represented by a difference equation as follows:
Outputs
After rearranging:
Inputs
Finally:
17
CEN352, Dr. Ghulam Muhammad King Saud University
Example 6
Identify non zero system coefficients of the following difference equations.
Solution:
b0 1, a1 0.25
Solution:
b0 1, b1 0.5
18
CEN352, Dr. Ghulam Muhammad King Saud University
System Representation Using Impulse Response
Impulse input with zero initial conditions
Impulse Response
Any input
y(n) = x(n) h(n) Convolution
19
CEN352, Dr. Ghulam Muhammad King Saud University
Example 7 (a)
Given the linear time-invariant system:
a.
Solution:
Therefore,
b.
20 CEN352, Dr. Ghulam Muhammad King Saud University
c.
Example 7 (b)
Solution:
a. Then
21
CEN352, Dr. Ghulam Muhammad King Saud University
Infinite!
Example 7 (b) contd.
b.
c.
Finite Impulse Response (FIR) system:
When the difference equation contains no previous outputs, i.e. a coefficients are zero. < See example 7 (a) >
Infinite Impulse Response (IIR) system:
When the difference equation contains previous outputs, i.e. a coefficients are not all zero. < See example 7 (b) >
22 CEN352, Dr. Ghulam Muhammad King Saud University
BIBO Stability
BIBO: Bounded In and Bounded Out
A stable system is one for which every bounded input produces a bounded output.
Let, in the worst case, every input value reaches to maximum value M.
Using absolute values of the impulse responses,
If the impulse responses are finite number, then output is also finite.
23 CEN352, Dr. Ghulam Muhammad King Saud University
Stable system.
BIBO Stability contd.
To determine whether a system is stable, we apply the following equation:
Impulse response is decreasing to zero.
24 CEN352, Dr. Ghulam Muhammad King Saud University
Example 8
Given a linear system given by: Which is described by the unit-impulse response: Determine whether the system is stable or not.
Solution:
Using definition of step function:
For a < 1, we know Therefore
25 CEN352, Dr. Ghulam Muhammad King Saud University
The summation is finite, so the system is stable.
Digital Convolution
The sequences are interchangeable. Commutative
x[n] * h[n] h[n] * x[n]
Convolution sum requires h(n) to be reversed and shifted. If h(n) is the given sequence, h(-n) is the reversed sequence.
26
CEN352, Dr. Ghulam Muhammad King Saud University
Reversed Sequence
Solution:
a.
27
CEN352, Dr. Ghulam Muhammad King Saud University
Convolution Using Table Method Example 9
Solution:
Length = 3
Length = 3
Convolution length = 3 +3 1 = 5
28 CEN352, Dr. Ghulam Muhammad King Saud University
Convolution Using Table Method Example 10
Solution:
Length = 3
Length = 2
Convolution length = 3 + 2 1 = 4
29 CEN352, Dr. Ghulam Muhammad King Saud University
Convolution Properties
Commutative:
Associative:
a[n] * b[n] b[n] * a[n]
(a[n] * b[n]) * c[n] a[n] * (b[n] * c[n])
Distributive:
a[n] * b[n] a[n] * c[n] a[n] * (b[n] c[n])
Associative
30 CEN352, Dr. Ghulam Muhammad King Saud University
Distributive
