Question 1

TELE4353 A cellular mobile communication system is using a

direct sequence code division multiple access
(CDMA) technique with coherent binary phase
Mobile and Satellite Communication shift keying (BPSK) modulation. The system is
Systems accommodating K users and the spreading
factor, G, for the system is 200.
Tutorial 5 (week 11-12) • With no other users present, the thermal noise is
the only noise in the receiver. If the SNR is 17
S2 2004 dB after the matched filter and dispreading
device, calculate the SNR before the dispreading
device.

Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 1 Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 2
Communication Systems Communication Systems

Question 1, cnt’d What is CDMA?

• If the system is accommodating 31 users, what is the bit • CDMA is a multiple access technique in
error probability for each user when SNR is 17 dB as which many users share the available
before. The probability of bit error for coherent BPSK is
given by
channel resource (code or power).
• The signal BW in CDMA system is much
1
(
Pe = erfc x = SNR
2
) 1
2π x
e− x
2

greater than the minimum required BW for

signal transmission.
Note that the multiple access interference could be • A pseudo noise (PN) random signal
approximated as additional additive white Gaussian noise
to the thermal noise.
converts the narrow band signal to wide
band noise like signal.
Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 3 Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 4
Communication Systems Communication Systems

What is direct sequence CDMA? Direct Sequence Spread

Spectrum System Model
• The original signal is multiplied by a very
m(t) – Message
large BW spreading (PN) signal in direct m(t)
Transmitter
p(t)
a(t) – Spreading Code
sequence CDMA system. n(t) p(t) – Spread Signal
a(t)
Channel
• The PN sequence has a much larger chip n(t) – AWGN Noise
d(t) r(t) – Received Signal
rate Rc than the signal data rate Rb. Receiver d(t) – Dispread Signal
r(t)

a(t)
r(t) = m(t) ⋅ a(t) + n(t)
d (t) = r(t) ⋅ a (t) = m (t) + a (t)n (t)
Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 5 Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 6
Communication Systems Communication Systems
 How does DS CDMA work? DS CDMA in Time and Frequency
• Each user has its own PN code which is Frequency Domain
Time Domain
orthogonal (near-orthogonal) to all other PN
codes. m(t) – Message 1 1 1
-1 -1
• The receiver needs to know the code used by the
transmitter.
a(t) – Spreading Code
• The power of other users at a receiver determines
the noise floor after signal dispreading and
demodulation. p(t) – Spread Signal

• Therefore it is necessary to control the power of

Band-limited Spread
each user in order to avoid excessive interference. Spectrum Signal

Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 7 Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 8
Communication Systems Communication Systems

Spectrum of SS Signal Spread Spectrum System

) = F{m(t) }


M(

M( ) 2
= N0Bm


P - signal power


p(t) = m(t) a(t)



Ba
Bm Bm G=
P( ) =M( ) ⊗A( ) Bm
) = F{ a(t) } P


A(

P( ) Bm
since Ba >> Bm P


SNRa =


Ba Bp ≈ Ba Ba G 2

M( )
P( ) = F{p(t) }


Bm Ba P N0 = k T P
G= SNRm =
 

2
Ba
Bm Bm
Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 9 Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 10
Communication Systems Communication Systems

Hiding Signal in Noise Solution to part 1

P( )
P
• The signals that we are dealing with have
AWGN
limited power P.
ω
• Effective noise power before spreading or
D(


) after dispreading is σ2 = N0Bm .

P
• Effective noise power after spreading or
AWGN N0 = k T
before dispreading is Gσ2 = N0Ba .
ω • The ratio of the noise power before and
Bm after dispreading is G = Ba / Bm= Tm / Ta.

Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 11 Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 12
Communication Systems Communication Systems
 Solution to part 1, cont’d Two SS Signals
M 1 (ω ) M 2 (ω )
P P P
• The SNR before spreading is SNRm =
N 0 Bm ω ω
Bm Bm
P
• The SNR after spreading is SNRa =
N 0 Ba P1 (ω ) P2 (ω )
N0 = k T ω
SNRm Ba SNRm
= =G SNRa =
SNRa Bm G Ba
~
M 1 (ω )
SNRm = 17dB P AWGN
SNRa = 17dB − 10 log(200) = −6dB
G = 200 Signal Interference N0 = k T ω

Bm Ba 2B a
Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 13 Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 14
Communication Systems Communication Systems

2 Signals in Noise Output Signal to Noise Ratio

Signal power - >P
r(t) = m1(t)a1(t) + m2 (t)a2 (t) + n(t)
Noise power -> = N 0 Bm

2

d1 (t) = r(t)a 1 (t)

P
Effective interference power ->
d1(t) = m1(t)a1(t) a1(t) + p2 (t)a1(t) + n(t)a1(t) G
P
-> 2+

d1(t) = m1(t) + p 2 (t)a1(t) + n(t)a1(t) Total noise power

G
P
Signal Interference Noise P 2 SNRm
SNRo = = 2
=
P P SNRm
2
+ 2
+ 1+
G G 2 G

Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 15 Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 16
Communication Systems Communication Systems

K SS Signals
k Signals in Noise M1(


) M2 (


) Mk (


)
P P P
•••
r(t) = m1(t)a1(t) + m2 (t)a2 (t) + ⋅ ⋅ ⋅ + mk (t)ak (t) + n(t) 



d1 (t) = r(t)a 1 (t) Bm Bm Bm

P1( ) • • •
 

d1(t) = m1(t)a1(t)a1(t) + p2 (t)a1(t) + ⋅ ⋅ ⋅ + pk (t)a1(t) + n(t)a1(t) P2 ( ) Pk ( )



N0 = k T 

d1(t) = m1(t) + p 2 (t)a 1(t) + ⋅ ⋅ ⋅ + p k (t)a 1(t) + n(t)a 1(t)

~ 
Ba
M1( )
Signal Interference Noise
P
AWGN
Signal
N0 = k T


Interference
Again we want to find output SNRo; -> BER
Bm Ba 2Ba
Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 17 Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 18
Communication Systems Communication Systems
 Output Signal to Noise Ratio Solution to part 2 of Q1
P SNRm
Signal power - >P SNRK = =
(k -1)P (k -1)
2
+ 1+ SNRm


-> 2 G G

Noise power
SNRm = 17 dB = 50.11
(k - 1)P
Effective interference power ->
G K = 31 SNRK = 5.89 = 7.7 dB
(k - 1)P
-> 2+

Total noise power G = 200

G
1
P SNRm Pe e −SNRK = = 2.56*10−4
SNRo = = 2π SNRK
(k - 1)P (k - 1) K = 5.89
+ 1+
2 SNR
SNRm
G G
Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 19 Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 20
Communication Systems Communication Systems

Question 2 Question 2, Cont’d

A convolutional coder with rate ½ is used for coding
• What is the output message from the
binary input sequence m = 10111. The following encoder multiplexer?
generator polynomials are used: g1 = 1+D2 and • Draw he encoder trellis diagram and
g2 = 1 + D + D2. Assume that the initial encoder indicate the encoder output sequence from
state is SI = 00 and two trailing bits Tr = 00 are the multiplexer on the diagram.
appended at the end of the message. • Use the Viterbi algorithm to find the most
• Draw the encoder block diagram and its state likely information sequence if the received
diagram. sequence (possibly erroneous) is given by:
r = 10, 01, 00, 11, 01, 11, 11

Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 21 Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 22
Communication Systems Communication Systems

Generator Sequences
Convolutional Encoder
Here n=2, k=1 and m=2. (2,1,2) convolutional code is specified
by the following sequences:
vl(1) v(1)
Input
g (1) = (1 0 1 ),
Path 1
v g ( 2 ) = (1 1 1 ).
Cl
C l-1 C l-2 Output
The input sequence is: c.
c vl(2)
v(2) The output sequence is:
Path 2

v (1) = c ∗ g (1) , vl(1) = cl ⊕ cl − 2

c = (1011100). v(1) = (1001011) v(2) = (1100101)
v ( 2) = c ∗ g (2) . vl(2) = cl ⊕ cl −1 ⊕ cl − 2
v = (11, 01, 00,10, 01,10,11)

Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 23 Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 24
Communication Systems Communication Systems
 Example of possible outputs
State Diagram
cl /vl(1) vl(2)
cl = 1 Each state is of the cl−1 cl−2 cl cl−1
form (Cl-1,Cl-2) 0 / 00
cl − 1 = 0 vl(1) = 1, vl(2) = 1,
00
vl(1) = cl ⊕ cl − 2 cl −2 = 0 0 / 11 1 / 11
vl(2) = cl ⊕ cl −1 ⊕ cl − 2 There are four states:
1 / 00
cl = 1 (0,0), (0,1), (1,1) and (1,0). 01 10
0 / 01
cl − 1 = 1 v(1)
l = 1, v
(2)
l =0 0 / 10 11
1 / 10

cl − 2 = 0
1 / 01

Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 25 Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 26
Communication Systems Communication Systems

Trellis Diagram Trellis Diagram

1/01
11 11 c = (1 0 1 1 1 0 0). v(1) = (1001011) v(2) = (1100101)
1/10 v = (11, 01, 00,10, 01,10,11)
0/1

1/01
10 10
0

11
0/

11 11 11
01

10
00 1/
1/

0/
10
10 10 10 10 10
1

01 01
1/1

0/
01

0/ 00
11
1/
1
1/1

01 01 01 01 01 0 /1
0/00 1
00 00
0/00 0/00 0/00 0/00 0/00 0/00 0/00
00 00 00 00 00 00 00 00

Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 27 Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 28
Communication Systems Communication Systems

Viterbi Algorithm Decoding stage 1, 2

• For each input sequence, we traverse a unique path in the
coder trellis and the coder output sequence is unique.
r = (10, 01)
• Due to the channel noise, we receive some bits with errors d=3
at the receiver. So the received sequence is no longer what 11 v 4 = (11,10), d (v4 , r ) = 3
was actually sent. 10
1/
d=2
• For decoding, we have to find the closest sequence that can 10 10 v 3 = (00,11), d ( v3 , r ) = 2
be outputed by the coder to the one that we have received.
0/
01

d=1
• If the input sequence has a length L bits, the input can take v 2 = (11, 01), d (v2 , r ) = 1
1
1/1

01
1

2L possibilities, so there are 2L different paths in the trellis

1/1

that can be taken, and we have to search among these 2L d=2 v1 = (00, 00), d ( v1 , r ) = 2
0/00 0/00
paths to find the closest to what we have received. 00 00 00
• Viterbi algorithm is an efficient way to find the closest path.
r= 10 01
Its complexity grows linearly with L.

Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 29 Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 30
Communication Systems Communication Systems
 Decoding, stage 3 Decoding, stage 4
d=3 d=2
3 3 d=3 d=3 3 3 d=2 d=2
3 3 3 3
01 01
11 11 11 11 11 11 11 11 11 11 11 11
11 11 11 11

0/1

0/1
10 2 d=1 10 1
d=3
2 2 d=1 1 1 d=3

0
2 d=1 1 d=3
10 10 10 10 10 10 10 10 10 10
10 10 10 10
00 00
0/

0/
1/ 1/
01

01
1 1 d=3 3 3 d=2 d=2
1 1 3 3
01 01 01 01 01 01 01 01 01
01 01 01
1

1
0/ 0/
1/1

1/1
11 11
2 2 d=2 2 2 d=3
2 2 2 2
00 00 00 00 00 00
0/00
00
00 00 00
0/00

00 00 00 11 11 11
00 11

Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 31 Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 32
Communication Systems Communication Systems

Decoding, stage 5 Decoding, stage 6

d=2 d=2 d=3 d=3
2 2 d=2 d=2 2 2 d=3 d=3
2 2 2 2
01 01
11 11 11 11 11 11 11 11 11 11 11 11
11 11 11 11
0/1

0/1
10 3
d=3 10 3
d=3
3 3 d=3 3 3 d=3
0

0
3 d=3 3 d=3
10 10 10 10 10 10 10 10 10 10
10 10 10 10
00 00
0/

0/
1/ 1/
01

01
2 2 d=3 d=3 3 3 d=3 d=3
2 2 3 3
01 01 01 01 01 01 01 01
01 01 01 01
1

0/ 0/
1/1

1/1

11 11
3 3 d=3 3 3 d=3
3 3 3 3
00 00 00 0/00 00 00 00 0/00
00 00 00 00

01 01 01 11 11 11
01 11

Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 33 Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 34
Communication Systems Communication Systems

Decoding, stage 7
Decoding stage 1-7
d=4 d=4
3 3 d=4 d=4
3 3
01
11 11 11 11 11 11 1
11 11 11 11
0/1

3
10 3
d=3 11 11 11 11
3 d=3
0

3 d=3 1 0
10 10 10 10 10 10 10 10
10 10 10 10 10
00 10
0/

1/
01

3 3 d=4 d=4 0
3 3 1
01 01 01 01 1
01 01 01
1

0/ 01 01
1/1

11 01 01 01
3 3 d=3 0
3 3
00 00 00 0/00
00 00 00
00 00 00 00 00 00
00
11 11 11 11
11 r= 01 01 00 11 01 11

Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 35 Tutorial 5- Week 11 – 12 TELE4353 Mobile and Satellite 36
Communication Systems Communication Systems