Nguyễn Công Phương

SIGNAL PROCESSING

Structures for Discrete – Time Systems

 Contents
I. Introduction
II. Discrete – Time Signals and Systems
III. The z – Transform
IV. Fourier Representation of Signals
V. Transform Analysis of LTI Systems
VI. Sampling of Continuous – Time Signals
VII.The Discrete Fourier Transform
VIII.Structures for Discrete – Time Systems
IX. Design of FIR Filters
X. Design of IIR Filters
XI. Random Signal Processing
s i tes.google.com/site/ncpdhbkhn 2
 Structures
for Discrete – Time Systems
1. Block Diagrams & Signal Flow Graphs
2. IIR System Structures
3. FIR System Structures
4. Lattice Structures
5. Structure Conversion, Simulation, &
Verification

s i tes.google.com/site/ncpdhbkhn 3
 Block Diagrams
& Signal Flow Graphs (1)
• Every practically realizable LTI system can be
described by a set of difference equations.
• A computational or implementation structure for a
discrete-time system is a pictorial block diagram
representation of the computational algorithm
using delays, adders, & multipliers.
• A system structure is for:
– Development of software that implements the system
on a general purpose computer or a special purpose
digital signal processor,
– Design & implementation of a hardware architecture
that can be used to implement the system using
discrete components or VLSI technology.
s i tes.google.com/site/ncpdhbkhn 4
 Block Diagrams
Ex. 1
& Signal Flow Graphs (2) y[n]
x [n ]
y[n ] = b0 x[n] + b1 x[ n − 1] − a1 y[ n − 1] b0

Y ( z ) = b0 X ( z ) + b1 X ( z ) z −1 − a1Y ( z ) z −1 z −1 z −1

b1 −a1
→ (1 + a1 z −1)Y ( z) = (b0 + b1 z −1 ) X ( z )
x[ n ]
y [n ]
Y ( z ) b0 + b1 z − 1
→ H (z) = = b0
X ( z ) 1 + a1 z − 1
z −1 z −1

b1 − a1
The transposition theorem: y [n ]
1. Reverse all branch directions, x [n ]
b0
2. Replace branch nodes by summing
nodes, & vice versa, & z −1 z −1
3. Interchange the input & output nodes.
− a1 b1

s i tes.google.com/site/ncpdhbkhn 5
 Block Diagrams
Ex. 1
& Signal Flow Graphs (3) y[n]
x [n ]
y[n ] = b0 x[n] + b1 x[ n − 1] − a1 y[ n − 1] b0

−1 z −1 z −1
Y ( z) b0 + b1 z
H ( z) = =
X ( z) 1 + a1 z −1 b1 −a1

v[n] = x[n ] − a1v[n − 1]

X (z)
→ V (z) =
1 + a1z −1

y[n ] = b0v[n ] + b1v[ n − 1] v [n ] y [n ]

x [n ]
→ Y ( z) = (b0 + b1 z −1 )V ( z ) b0

z −1 z −1
Y ( z ) b0 + b1 z −1
→ H (z) = = − a1 b1
X ( z ) 1 + a1 z −1
s i tes.google.com/site/ncpdhbkhn 6
 Structures
for Discrete – Time Systems
1. Block Diagrams & Signal Flow Graphs
2. IIR System Structures
3. FIR System Structures
4. Lattice Structures
5. Structure Conversion, Simulation, &
Verification

s i tes.google.com/site/ncpdhbkhn 7
 IIR System Structures (1)
3 structures:
1. A direct form which is obtained directly from
the system function & has two variations,
2. A cascade form which is obtained by
expressing the system function as a product
of second-order sections, &
3. A parallel form which is obtained by
expressing the system function as a sum of
second-order sections.
s i tes.google.com/site/ncpdhbkhn 8
 IIR System Structures (2)
N
Direct Form I Structure
M
y[n ] = −  a k y[n − k ] +  bk x[n − k ] x [n ]
b0 s[ n ]
y[n]
k =1 k =0
z −1 z −1
b1 − a1
y[n − 1]
M
x [ n − 1]
Y (z)  bk z −k

z −1
H (z) = = k =0 z −1
N b2 −a 2
X (z) x [n − 2] y[n − 2]
1 +  ak z − k
k =1

M
S( z ) M
s[n ] =  bk x[n − k ] → H1 ( z ) = =  bk z − k
k =0 X ( z ) k =0
N
Y ( z) 1
y[n ] = −  a k y[n − k ] + s[n] → H2 (z) = = N
S( z)
k =0
1 +  a k z −k
k =1 M

Y (z) Y (z) S (z) b z k

−k

→ H (z) = = . = H 2 ( z) H 1 ( z ) = k=0
N
X ( z ) S ( z ) X ( z)
1 +  a k z −k
k =1
s i tes.google.com/site/ncpdhbkhn 9
 IIR System Structures (3)
Transposed Direct Form I Structure
N M
y[n ] = −  a k y[n − k ] +  bk x[n − k ] x [n ]
b0 s[ n ]
y[n]
k =1 k =0
z −1 z −1
b1 − a1
y[n − 1]
M
x [ n − 1]
Y (z)  bk z −k

z −1
H (z) = = k =0 z −1
N b2 −a 2
X (z) x [n − 2] y[n − 2]
1 +  ak z − k
k =1

w[ n ] b0
x [n ] y[n]
z −1 z −1
−a1 b1

z −1 z −1
− a2 b2

s i tes.google.com/site/ncpdhbkhn 10
 IIR System Structures (4)
Direct Form II Structure
b0 s[ n ]
x [n ] y[n]
H ( z) = H 2 ( z ) H 1( z )
z −1 z −1
1 M b1 − a1
= N b z k
−k x [ n − 1] y[n − 1]
1 +  ak z − k k=0
z −1
b2 −a 2
z −1
k =1 x [n − 2] y[n − 2]

H ( z ) = H1 ( z) H 2 ( z ) → Y ( z ) = H1 ( z ) H 2 ( z) X ( z) = H 1( z )W ( z )
1
W ( z ) = H 2 ( z) X ( z ) = N
X ( z)
1 +  ak z −k
k =1

N N
→ W ( z ) +  ak z W ( z ) = X ( z )
−k
→ w[n ] = −  a k w[n − k ] + x[n ]
k =1 k =1
M M
Y ( z ) =  bk z W ( z )
−k
→ y[ n] =  bk w[n − k ]
k =0 k =1
s i tes.google.com/site/ncpdhbkhn 11
 IIR System Structures (5)
Direct Form II Structure
b0 s[ n ]
x [n ] y[n]
H ( z) = H 2 ( z ) H 1 ( z )
z −1 z −1
1 M b1 − a1
= N b z k
−k x [ n − 1] y[n − 1]
1 +  ak z − k k =0
z −1
b2 −a 2
z −1
k =1 x [n − 2] y[n − 2]

x [n ] w[ n ] b0 y[n]
N
w[n ] = −  ak w[n − k ] + x[n ]
z −1
k =1 − a1 b1
M
w[ n − 1]
y[n ] =  bk w[n − k ] z −1
k =1 −a 2 b2
w[ n − 2]

s i tes.google.com/site/ncpdhbkhn 12
 IIR System Structures (6)
Direct Form
x [n ] b0 s[ n ] y[n] x [n ] w[ n ] b0 y[n]

z −1 z −1 z −1 z −1
b1 − a1 −a1 b1

z −1 z −1 z −1 z −1
b2 −a 2 − a2 b2

Direct form I M Transposed direct form I

b z k
−k

H (z) = k =0
N
1 +  a k z −k
Direct form II k =1
Transposed direct form II
x [n ] w[ n ] b0 y[n] x [n ] b0 y[n]

z −1 z −1
−a1 b1 b1 −a1

z −1 z −1
−a 2 b2 b2 − a2
s i tes.google.com/site/ncpdhbkhn 13
 IIR System Structures (7)
Ex. 2
Direct Form
2 + z −1 + 0.3z −2 − 0.8 z −3 + 6 z −4
H (z) =
1 − 2 z −1 + 4 z −2 − 3 z −3 + 0.5 z −4

s i tes.google.com/site/ncpdhbkhn 14
 IIR System Structures (1)
3 structures:
1. A direct form which is obtained directly from
the system function & has two variations,
2. A cascade form which is obtained by
expressing the system function as a product
of second-order sections, &
3. A parallel form which is obtained by
expressing the system function as a sum of
second-order sections.
s i tes.google.com/site/ncpdhbkhn 15
 IIR System Structures (8)
Cascade Form
M M1 M2

b z k
−k
∏ (1 − z z )∏ (1 − z z
k
−1
k
−k
)(1 − zk* z −1 )
H (z) = k =0
N
= b0 k =1
N1
k =1
N2
1 +  a k z −k
k =1
∏ (1
k =1
− pk
z −1
) ∏ (1 − p
k =1
k
z −k
)(1 − pk
* −1
z )

Bk 0 + Bk 1 z −1 + Bk 2 z −2
K
= G∏ −1
k =1 1 + Ak 1 z + Ak 2 z −2

x [n ] G B10 B20 y[n]

z −1 z −1
B11 − A11 B21 − A21

z −1 z −1
B12 − A12 B22 − A22

s i tes.google.com/site/ncpdhbkhn 16
 IIR System Structures (9)
Ex. 2
Cascade Form
2 + z −1 + 0.3z −2 − 0.8 z −3 + 6 z −4
H (z) =
1 − 2 z −1 + 4 z −2 − 3 z −3 + 0.5 z −4
1 + 2.106 z −1 + 2.1119 z −2 1 − 1 .606 z −1 + 1.4205 z −1
= 2× −1 −2
×
1 − z + 0.1771z 1 − z −1 + 2.8229 z −2

2.106 1 –1.606 1

2.1119 –0.1771 1.4205 –2.8229

s i tes.google.com/site/ncpdhbkhn 17
 IIR System Structures (1)
3 structures:
1. A direct form which is obtained directly from
the system function & has two variations,
2. A cascade form which is obtained by
expressing the system function as a product
of second-order sections, &
3. A parallel form which is obtained by
expressing the system function as a sum of
second-order sections.
s i tes.google.com/site/ncpdhbkhn 18
 IIR System Structures (10)
Parallel Form
M

b z k
−k
M −N K
Bk 0 + Bk 1 z −1
H (z) = k =0
N
= Cz k
−k
+
k =1 1 + Ak 1 z
−1
+ Ak 2 z −2
1 +  ak z −k k =0

k =1

s i tes.google.com/site/ncpdhbkhn 19
 IIR System Structures (11)
Ex. 2
Parallel Form
2 + z −1 + 0.3z −2 − 0.8 z −3 + 6 z −4 1.69 + 2.70 z −1 − 11.69 + 12. 30 z −1
H (z) = −1 −2 −3 −4
= 12 + −1 −2
+
1 − 2 z + 4 z − 3 z + 0.5 z 1 − z + 2.82 z 1 − z −1 + 0.18 z −2

12
x [n ] y [n ]
1. 69

z −1
2.70 1

z −1
−2.82

−11.69

z −1
12.30 1

z −1
−0.18

s i tes.google.com/site/ncpdhbkhn 20
 Structures
for Discrete – Time Systems
1. Block Diagrams & Signal Flow Graphs
2. IIR System Structures
3. FIR System Structures
4. Lattice Structures
5. Structure Conversion, Simulation, &
Verification

s i tes.google.com/site/ncpdhbkhn 21
 FIR System Structures (1)
M
y[n ] =  bk x[n − k ]
k =0

bn , n = 0, 1,..., M
h[n] = 
0, otherwise
Y (z) M M
H (z) = =  bn z =  h[n ]z − n
−n

X ( z ) n =0 n =0

1. Direct form
2. Cascade form
3. Direct form for linear-phase FIR systems
4. Frequency-sampling form
s i tes.google.com/site/ncpdhbkhn 22
FIR System Structures (2)
Direct Form

s i tes.google.com/site/ncpdhbkhn 23
FIR System Structures (3)
Cascade Form
M K
H ( z ) =  h[n ]z −n
= G ∏ (1 + Bɶk 1 z −1 + Bɶ k 2 z −2 )
n =0 k =0

s i tes.google.com/site/ncpdhbkhn 24
 FIR System Structures (1)
M
y[n ] =  bk x[n − k ]
k=0

bn , n = 0, 1,..., M
h[n] = 
0, otherwise
Y (z) M M
H (z) = =  bn z =  h[n ]z − n
−n

X ( z ) n =0 n =0

1. Direct form
2. Cascade form
3. Direct form for linear-phase FIR systems
4. Frequency-sampling form
s i tes.google.com/site/ncpdhbkhn 25
 FIR System Structures (4)
Direct Form for Linear-Phase FIR Systems
M
y[n ] =  h[k ] x[n − k ]
k =0

M
−1
2
M   M M
=  h[k ]x[n − k ] + h   x n −  +  h[k ]x[n − k ]
k =0  2  2  k = M +1
2

M M
−1 −1
2
M   M 2
= 
k =0
h[ k ] x[ n − k ] + h x
 2   n −  +  h[ M − kk ] x[n − M + k ]
2  k =0

M
−1
2
M   M
=  h[ k ] ( x [ n − k ] + x[ n − M + k ] ) + h x
 2   n −
k =0 2 

s i tes.google.com/site/ncpdhbkhn 26
 FIR System Structures (5)
Direct Form for Linear-Phase FIR Systems
M
−1
M 2
M   M
y[n ] =  h[k ]x[n − k ] =  h[ k ] ( x[ n − k ] + x [ n − M + k ] ) + h x
 2   n −
k =0 k =0 2 

s i tes.google.com/site/ncpdhbkhn 27
 FIR System Structures (6)
Ex. 3 Direct Form for Linear-Phase FIR Systems
H ( z ) = 5 − 10 z − 1 + 5 z −2 − 20 z −3 + 35 z −4 − 20 z − 5 + 5 z − 6 − 10 z − 7 + 5 z −8

s i tes.google.com/site/ncpdhbkhn 28
 FIR System Structures (7)
Ex. 3 Direct Form for Linear-Phase FIR Systems
H ( z ) = 5 − 10 z − 1 + 5 z −2 − 20 z −3 + 35 z −4 − 20 z − 5 + 5 z − 6 − 10 z − 7 + 5 z −8

s i tes.google.com/site/ncpdhbkhn 29
 FIR System Structures (1)
M
y[n ] =  bk x[n − k ]
k =0

bn , n = 0, 1,..., M
h[n] = 
0, otherwise
Y (z) M M
H (z) = =  bn z =  h[n ]z − n
−n

X ( z ) n =0 n =0

1. Direct form
2. Cascade form
3. Direct form for linear-phase FIR systems
4. Frequency-sampling form
s i tes.google.com/site/ncpdhbkhn 30
 FIR System Structures (8)
Frequency-Sampling Form

1 − z− N N −1
H [k ]
H (z) =
N

k =0
j
2π
k
, H [k ] = H ( z ) z = e j 2 π k / N
1 − z −1e N

1− z −N  H [0] H [ N / 2] K 
=
N

 1 − z −1
+
1 + z −1
+ 
k =1
2 H [ k ] H k ( z ) 


cos(∠H [k ]) − z −1 cos(∠H [k ] − 2π k / N )
Hk (z) =
1 − 2 cos(2π k / N ) z −1 + z −2

s i tes.google.com/site/ncpdhbkhn 31
FIR System Structures (9)
Frequency-Sampling Form

s i tes.google.com/site/ncpdhbkhn 32
 Structures
for Discrete – Time Systems
1. Block Diagrams & Signal Flow Graphs
2. IIR System Structures
3. FIR System Structures
4. Lattice Structures
a) All-Zero Lattice Structure
b) All-Pole Lattice Structure
5. Structure Conversion, Simulation, &
Verification

s i tes.google.com/site/ncpdhbkhn 33
Lattice Structure

s i tes.google.com/site/ncpdhbkhn 34
 Structures
for Discrete – Time Systems
1. Block Diagrams & Signal Flow Graphs
2. IIR System Structures
3. FIR System Structures
4. Lattice Structures
5. Structure Conversion, Simulation, &
Verification

s i tes.google.com/site/ncpdhbkhn 35
 Structure Conversion,
Simulation, & Verification (1)

s i tes.google.com/site/ncpdhbkhn 36
 Structure Conversion,
Simulation, & Verification (2)
• Verification of the converted structure can be obtained
in many different ways. The simplest is to:
– (a) convert the given structure to direct form I,
– (b) use the filter function to implement this structure,
– (c) excite the structure in (b) and the structure to be tested
with the same input and compare the corresponding
outputs.
• Typical test inputs include the unit step, sinusoidal
sequences, or random sequences.
• Another way is to obtain a well-known response like an
impulse or a step response from one structure and
compare it with that from another structure.

s i tes.google.com/site/ncpdhbkhn 37