Module 5 Convolutional Codes

Manjunatha. P
manjup.jnnce@gmail.com

Professor
Dept. of ECE

J.N.N. College of Engineering, Shimoga

November 13, 2024

 Module 5 Convolutional Codes Convolutional Codes

2022 March For a (2,1,2) convolutional encoder with

g (1) = [1 1 1] g (2) = [1 0 1]

i. Draw the state transition table

ii. Draw the State diagram
iv. Draw the Trellis diagram
iii. Draw the code tree.

Solution:
(n,k,m) n=number of output=2, k=number of input=1, and m = number of memory
registers=2
V1

Input
FF FF

V2

Figure 1: Encoder for g (1) = [1 1 1] g (2) = [1 0 1]

Manjunatha. P (JNNCE) Module 5 Convolutional Codes November 13, 2024 2 / 13
 Module 5 Convolutional Codes Convolutional Codes

The convolutional encoder consists of 2 Flip-flops.

The content of the Flip-flops are 00 01 10 11.
Table 1: State transition table
g (1) (D) = 1 + D + D 2 g (2) (D) = 1 + D 2

Input Present State Next State Output

Binary description State 0 00 00 00
00 a 1 00 10 11
01 c 0 01 00 11
10 b 1 01 10 00
11 d 0 10 01 10
1 10 11 01
0 11 01 01
1 11 11 10

Manjunatha. P (JNNCE) Module 5 Convolutional Codes November 13, 2024 3 / 13

 Module 5 Convolutional Codes Convolutional Codes

Table 2: State transition table State b

10
0(10)
g (1) (D) = 1 + D + D 2 g (2) (D) = 1 + D 2 1(11)

1(00)
Input Present State Next State Output
1(01)
0 00 00 00
1 00 10 11 0(00)
State a
00
0(11) State c
01

0 01 00 11
1 01 10 00
0(01)
0 10 01 10
1 10 11 01 1(10)
State d
11
0 11 01 01
1 11 11 10 Figure 2: State graph

Binary description State

00 a
01 c
10 b
11 d

Manjunatha. P (JNNCE) Module 5 Convolutional Codes November 13, 2024 4 / 13

 Module 5 Convolutional Codes Convolutional Codes

Table 3: State transition table a

00 00 00 00 00 00 00 00
g (1) (D) = 1 + D + D 2 g (2) (D) = 1 + D 2 11 11 11 11 11
11

Input Present State Next State Output b 11 11 11 11 11 11

0 00 00 00 00 00 00

1 00 10 11 10
c
0 01 00 11 01 01 01 01 01
1 01 10 00 01 01 01 01 01 01
0 10 01 10 d 10 10 10 10 10
1 10 11 01 Level J= 0 1 2 3 4 5 L-1 L L+1 L+2

0 11 01 01
1 11 11 10 Figure 3: Trellis graph

Binary description State

00 a
01 c
10 b
11 d

Manjunatha. P (JNNCE) Module 5 Convolutional Codes November 13, 2024 5 / 13

 Module 5 Convolutional Codes Convolutional Codes

00
00 a
11
00 a
10
11 b
01
00 a
Binary description State 10 c
11

11 00
00 a b
01
01 d
01 c 10
00 a
00
10 b 11 a
11
10
11 d c
10
00 b
01
11 b
11
01 c
00
01
Table 4: State transition table
0
d
10 d
01

g (1) (D) = 1 + D + D 2 g (2) (D) = 1 + D 2 10

00
00 a
11
11 a
Input Present State Next State Output 1
11 b
10

01
0 00 00 00 10 c
11
10 c
1 00 10 11 00
00
b
0 01 00 11 01 d
01

10
1 01 10 00 11
b
00
11 a
0 10 01 10 01 c
11

10
1 10 11 01 00 b
01
01
0 11 01 01 d
01 c
11

00
1 11 11 10 10
d
01
10 d
10

Figure 4: Tree graph

Manjunatha. P (JNNCE) Module 5 Convolutional Codes November 13, 2024 6 / 13
 Module 5 Convolutional Codes Convolutional Codes

For a (3,1,2) convolutional encoder with

g (1) = [1 1 0] g (2) = [1 0 1] g (3) = [1 1 1]

i. Draw the encoder diagram

ii. Draw the state transition table
iii. Draw the State diagram
iv. Draw the Trellis diagram

Solution:
(n,k,m) n=number of output=3, k=number of input=1, and m = number of memory
registers=2
V (0)

Input U
SR SR V (1)

Output

V (2)

Figure 5: Encoder for g (1) = [1 1 0] g (2) = [1 0 1] g (3) = [1 1 1]

Manjunatha. P (JNNCE) Module 5 Convolutional Codes November 13, 2024 7 / 13
 Module 5 Convolutional Codes Convolutional Codes

G(D) = [1 + D, 1 + D 2 , 1 + D + D 2 ]

Table 5: State transition table for the (3,1,2)encoder

V (0)

Input Present State Next State Output

Input U
0 00 00 000
SR SR V (1) 1 00 10 111
Output 0 01 00 011
1 01 10 100
V (2) 0 10 01 101
1 10 11 010
0 11 01 110
Figure 6: R=1/3 convolutional encoder with 1 11 11 001
memory m=2

S1

0(101)
1(010)

1(111)
Tuple State 1(100)
00 s0
0(011)
01 s2 0(000) S0 S2
10 s1
11 s3

0(110)

1(001) S3

Manjunatha. P (JNNCE) Module 5 Convolutional Codes November 13, 2024 8 / 13

 Module 5 Convolutional Codes Convolutional Codes

001 001 001

S3 S3 S3 S3

110 110 110

110
Table 6: State transition table
010 010 010 010
g (1) (D) = 1 + D + D 2 g (2) (D) = 1 + D 2
S1 S1 S1 S1 S1

Input Present State Next State Output 101 101 101 101 101
0 00 00 000 100 100
1 00 10 111 S2 S2 S2 S2 S2
0 01 00 011 011 011 011
011 011
1 01 10 100

111

111
111

111
111
0 10 01 101
S0 S0 S0 S0 S0 S0 S0 S0
1 10 11 010 000 000 000 000 000 000 000
0 11 01 110 0 1 2 3 4 5 6 7
Time units
1 11 11 001

Figure 7: Trellis graph

Binary description State

00 s0
01 s2
10 s1
11 s3

Manjunatha. P (JNNCE) Module 5 Convolutional Codes November 13, 2024 9 / 13

 Module 5 Convolutional Codes Convolutional Codes

Problem 9.41 Proakis A convolutional code is described by

g1 = [1 0 0] g2 = [1 0 1] g3 = [1 1 1]

a. Draw the encoder diagram and find the output using time domain for the input sequence (1 1 1 0 1).
b. Draw the state-transition diagram.
c. Draw the trellis diagram.
d. If the received sequence is r =(110, 110, 110, 111,010, 101, 101), using the Viterbi algorithm find the transmitted bit
sequence.
Solution:

a.

V (0)
 
111 001 011
Input U
 111 001 011 
V (1) G= 
 111 001 011 
Output
111 001 011
V (2)

111 001 011

 
 111 001 011 
V = UG = (1 1 1 0 1)  111 001 011  = (111 110 101 010 100 100 001 011 )
 
 111 001 011 
111 001 011

Manjunatha. P (JNNCE) Module 5 Convolutional Codes November 13, 2024 10 / 13

 Module 5 Convolutional Codes Convolutional Codes

b.
V (0) Table 7: State table g1 = [1 0 0] g2 = [1 0 1] g3 = [1 1 1]

Input Present State Next State Output

Input U 0 00 00 000
V (1) 1 00 10 111
Output 0 01 00 011
1 01 10 100
0 10 01 001
V (2) 1 10 11 110
0 11 01 010
1 11 11 101
Figure 8:
g1 = [1 0 0] g2 = [1 0 1] g3 = [1 1 1]

S1

0(101)
1(010)

1(111)
1(100)

Tuple State 0(000) S0

0(011)
S2
00 a
01 b
10 c
11 d
0(110)

1(001) S3

Manjunatha. P (JNNCE) Module 5 Convolutional Codes November 13, 2024 11 / 13

 Module 5 Convolutional Codes Convolutional Codes

S1

0(101)
1(010)

1(111) Tuple State

1(100)
00 a
c. 0(000) S0
0(011)
S2 01
10
b
c
11 d

0(110)

1(001) S3

000 000 000 000 000

a
011 011 011
111 111 111 111 111

b
100 100 100
001 001 001 001
c 011
110 110 110 110
010 010 010
d
101 101 101

Steady State

Figure
Manjunatha. P 10:
(JNNCE) Trellis diagram for rateModule
1/3 m=2 convolutional
5 Convolutional Codes code (3,1,2) November 13, 2024 12 / 13
 Module 5 Convolutional Codes Convolutional Codes

110 110 110 111 010 101 100

2 4 6 3 4 6 7
000 000 000 000 000 000 000
a X
011 011 011 X 011 011
111 111 111 111 111
X
4 2 5 4 6
b X
001 100 100 X 100 001
001
X 001 001
4 5
1 3 5
c 011
110 110 110 110
010 010
010 X 010
1 3 4 5
d
101 101 101 101

Steady State

The path traced by the Viterbi is (111, 110, 010, 011, 000, 000, 000) and the
decoded sequence is (1 1 0 0 0 0 0)
Manjunatha. P (JNNCE) Module 5 Convolutional Codes November 13, 2024 13 / 13