Helsinki University of Technology
Department of Electrical and Communication Engineering
WCDMA Simulator with Smart Antennas
Hong Zhang
Communication laboratory
Supervisor: Professor Seppo J. Halme
Instructor: M. Sc. Adrian Boukalov
10 January,2002
Seminar of Master Thesis
1
Helsinki University of Technology
Department of Electrical and Communication Engineering
Outline
1. Background
2. Different approach in WCDMA system modelling
3. Spreading in WCDMA
4. RAKE Receiver and Multiuser Detection
5. Smart Antenna in WCDMA
6. Simulation Results
7. Conclusion
10 January,2002
Seminar of Master Thesis
2
Helsinki University of Technology
Department of Electrical and Communication Engineering
1. Background
The goal of 3G
to provide a wide variety of communication
services and high speed data access.
WCDMA
radio access technology for 3G
The increasing demand of high
capacity
To provide high capacity
technique
Spreading
Smart antenna
RAKE receiver
Multiuser detection
tool
Simulation
10 January,2002
Seminar of Master Thesis
3
Helsinki University of Technology
Department of Electrical and Communication Engineering
2. Different approach in
WCDMA system modelling (1)
CDMA System Modelling
With linear
algebra knowledge
Synchronous CDMA
complexity
Modeling with AWGN channel
Modeling with channel fading
Asynchronous CDMA
Modeling with single path
Modeling with multipath
Modeling with smart antenna
10 January,2002
Seminar of Master Thesis
4
Helsinki University of Technology
Department of Electrical and Communication Engineering
2. Different approach in WCDMA system modelling (2):
Mobile radio channel
Linear time-variant system
N 1
h (t, , θ) = 
βj (t)(  - j ) ( θ - θj)
j 0
β( ,t)
vmax max<<1
Quasi-stationary
Multiple antennas in the receiver
βt()
N 1
Uncorrelated scatters
GWSSUS
 βi(t)(  - i )
h (t, , θ) = 
β (t)(  - j ) a( θ - θj)
j 0 j
i
Ray tracing model
Sampled
N 1
Tapped delay line
 βj(t) (  - jT)
The tapped delay line model
j 0
Ws max<<1
Narrowband model
T
β0(t)
β0
T
β1
T
β N-1
where βj (t) is the complex amplitude, j is path delay and θj is
Direction Of Arrival, a( θ - θj) is the steering vector
GWSSUS: Gaussian Wide Sense Stationary Uncorrelated Scatters
DOA: Direction Of Arrival
10 January,2002
Seminar of Master Thesis
5
Helsinki University of Technology
Department of Electrical and Communication Engineering
2. Different approach in WCDMA system modelling (3):
the fundamental structure
Block diagram of the mobile transmitter for user k
spreader
source of
information
bk
encoder/
interleaver
dk
transmitter
filter
xk
IF-RF
upconverter
D/A
xk(t)
PN code
Block diagram of the base station receiver
Despreader unit 1
sink 1

b
multiuser
detector
unit
PN code
yMF

receiver
front – end
Despreader unit K
sink K
r (t)
PN code
10 January,2002
Seminar of Master Thesis
6
Helsinki University of Technology
Department of Electrical and Communication Engineering
2. Different approach in WCDMA system modelling (4): Discrete time base
band uplink model for asynchronous CDMA

r=
The received signal
L 1
K
M
i 0
k 1
m 1
  
Discrete time model base band uplink
model for asynchronous CDMA system
over a mobile radio channel
 OiNQ



cˆk ,m (i ) d k (i )  ŝ k ,m (i)  + n=SCd+n


 O( L i 1) NQ 


 The
output after matched
filtering and correlating
s1(i)
u1
p
c1,1(i) ... c1,M(i)
p
c2,1(i) ... c2,M(i)
d1(i)
y =CH SH S C d + CH SH n
s2(i) u1
 O

 k, m



=  sˆ k (i )

O
 Mˆ  

k
,
m


Where
sˆ k , m (i )
d k (i ) :
the ith data symbol transmitted by user k
d2(i)
.
.
.
sˆ k (i ) : the zero-padded spreading sequence for symbols generated by user k
cˆ k ,m (i ) : the channel coefficients over the mth multipath for the ith symbol generated by user
L : the total amount of the data symbol sent by every active user
K : the total amount of active users sending data to the BTS
M̂ : the maximum delay spread normalised to sampling interval
M : the amount of resolvable multipath components per user data sequence
C: channel matrix
S: the matrix of received spreading waveform
n: Additive White Gaussian Noise
10 January,2002
k.
sK(i) u1
.
.
.
.
.
.
p
r
n
cK,1(i) ... cK,M(i)
dK(i)
Seminar of Master Thesis
7
Helsinki University of Technology
Department of Electrical and Communication Engineering
2. Different approach in WCDMA system modelling (5): Discrete time
base band uplink model for asynchronous CDMA with smart antenna
 The
phase difference of the received
signal between adjacent antenna elements
ф=
 The
The ULA with the direction of arrival
(α, φ) indicated
2d cos
z

’
φ
received signal
d
r = S Ψ C0 d +n
 The
B-1
output after matched filtering
and correlating
y=
D
θ
B
wavefor
m
y
α
1
x
C0H ΨHSH r = C0H ΨHSH S ΨC0 d + C0H ΨHSH n
where Ψ is the steering matrix
10 January,2002
Seminar of Master Thesis
8
Helsinki University of Technology
Department of Electrical and Communication Engineering
3. Spreading in WCDMA (1)
• Pseudo Random (PN) sequence:
a bit stream of ‘1’s and ‘0’s occurring
randomly, or almost randomly, with some
unique properties.
Linear shift register
an
an-1
c1
an-2
c2
an-r
c3
cr
an = c1 an-1 + c2 an-1 + ... + cr an-
10 January,2002
Seminar of Master Thesis
9
Helsinki University of Technology
Department of Electrical and Communication Engineering
3. Spreading in WCDMA (2) : Spreading and scrambling at the uplink
Spreading: to multiply the input information bits by a PN code and get processing gain, the
chip level signal’s bandwidth is much wider than that of input information bits.
It maintains the orthogonality among different physical channels of each user.
Scrambling: to separate the signals from the different users. It doesn’t change the signal
bandwidth. Each user has a unique scrambling code in the system.
Channelization codes
(Walsh/OVSF)
• Uplink spreading
(Cd )
DPDCH
Scrambling
codes (Gold)
(Csc)
cos ( ωt)
P(t)
and modulation
DPCCH
P(t)
j
sin ( ωt)
Channelization codes
(Walsh/OVSF) (Cc)
bit rate
chip rate
chip rate
WCDMA
Selecting codes
an interference limited system
high autocorrelation low cross correlation
10 January,2002
Seminar of Master Thesis
Suppressing
interference
10
Helsinki University of Technology
Department of Electrical and Communication Engineering
3. Spreading in WCDMA (3) : Walsh-Hadamard code and Gold code
• Walsh-Hadamard code
Purpose: spreading
 Generation: code tree

C 2,1 = ( 1 , 1 )
C 4,1 = ( 1 , 1 , 1 , 1 )
C 4, 2 = ( 1 , 1 , -1 , -1 )
C 1,1 = ( 1 )
C 2,2 = ( 1 , 1 )
C 4, 3 = ( 1 , -1 , 1 , -1 )
C 4,4 = ( 1 , -1, -1, 1 )
SF=1
SF=2
SF=4
• Gold code
Purpose: scrambling
 Generation: modulo-2 sum 2 m-sequences

10 January,2002
Seminar of Master Thesis
11
Helsinki University of Technology
Department of Electrical and Communication Engineering
4. RAKE Receiver and Multiuser Detection (1): RAKE receiver
r
RAKE receiver:
to collect the signal energy from
different multipath components
and coherently combine the signal
Output : y = CH SH r
1
W
Finger 1
1
W
Finger M
ck,1(i)*
ck,M(i)*
Correlator
Correlator
sk(i)*
sk(i)*
SNR

Optimal
for single user system
T
o
• Combining methods

()
T
o
y
 Selection Combining
 Maximal Ratio Combining (MRC)
 Equal Gain Combining (EGC)
10 January,2002
Seminar of Master Thesis
12
()
Helsinki University of Technology
Department of Electrical and Communication Engineering
4. RAKE Receiver and Multiuser Detection (2): Multiuser Detection
WCDMA
Optimal Detector
User 1 RAKE
MAI
High complex
MUD
design and
analysis the
digital
demodulation in
the presence of
MAI
r(t)
User 2 RAKE

User K RAKE

Viterbi Algorithm
Multiple access
d 1 (i )

d 2 (i )


d K (i )
Sub Optimal
Detector
LMMSE: Linear Minimum Mean Square Error
MUD: multiuser detection
MAI: multiple access interference
10 January,2002
Seminar of Master Thesis
13
Helsinki University of Technology
Department of Electrical and Communication Engineering
4. RAKE Receiver and Multiuser Detection (3): Multiuser Detection

T
y1
d 1 (i )
-
Decorrelating Detector
d̂ Dec = R-1y (synchronous)

+
0
s1(t)
r(t)
d̂ Dec = [RT [1] z + R [0] + R [1] z-1 ]-1 y

T

0
s2(t)
(asynchronous)

T
y2
+

d 2 (i )
0
Noise
Decorrelating detector for 2 synchronous users
Sub
Optimal
Detector
LMMSE
d̂ Dec = (R+ σ2I)-1 y (synchronous)
d̂ Dec = (RT [1] z + R [0] + R [1] z -1 + σ2I )-1 y
(asynchronous)
Varying channel
LMMSE: Linear Minimum
Mean Square Error
MUD: multiuser detection
MAI: multiple access
interference
10 January,2002
Adaptive MMSE algorithmRLS algorithm with adaptive memory
Seminar of Master Thesis
n
J(k) =

n i
(| ζ(k)|2 )
i 1
14
Helsinki University of Technology
Department of Electrical and Communication Engineering
5. Smart Antenna in WCDMA (1)
Smart Antenna consists of antenna array, combined with signal
processing in space (or time) domain
• switched beam antenna array
Broad-band beam-former structure
Steering Delay
τ1 (φi,θi)
Desired signal
r1
T
w11
Interfering signal
T
w12
w1M

τB (φi,θi)
Type
• adaptive antenna array
rB
wB1
y(t)
T
T
wB2
Desired signal
wBM
Weight
control
Error
signal
+
Reference signal
Interfering signal
B
y(t) =  wn rn (t)
n 1
10 January,2002
Seminar of Master Thesis
-
15
Helsinki University of Technology
Department of Electrical and Communication Engineering
5. Smart Antenna in WCDMA (2): Beamforming schemes
Conventional
Beamforming
wc = (1 /B ) s
Statistically
Optimum
Beamforming
MMSE
Max SNR
LCMV
MMSE: mimimize mean square error
LCMV: Linearly Constrained Minimum Variance
RLS: recursive least squares
w = R-1 p
Rn-1Rs w = λmax w
w = R-1c[cHR-1c]-1g
G(k) =
Adaptive
Beamforming
RLS algorithm
with adaptive
memory
1 P ( k  1) r (k )
1  1 r H ( k ) P (k  1) r (k )
ζ(k) = d(k) - ŵ
H
(k-1) r(k)
w(k)= ŵ (k-1) + G(k) ζ *(k)
P(k) = λ-1 P(k-1) - λ-1 G(k) rH (k) P(k-1)
λ(k)= λ(k-1)+α Re[ˆ H ( k  1) r(k) ζ *(k)]] 
S(k) = λ-1[I - G(k) rH (k)] S(k-1) [I - r (k)GH(k)]+ λ-1 G(k) GH(k)- λ-1 P(k)
 (k ) = [I - G(k) rH (k)] ˆ H ( k  1) + S(k) r(k) ζ *(k)
10 January,2002
Seminar of Master Thesis
16
Helsinki University of Technology
Department of Electrical and Communication Engineering
6. Simulation (1)
• Simulation block diagram of transmitter in WCDMA uplink
spreading
d1(i)
( user 1)
d2(i)
( user 2)
Modulation
(BPSK)
bit
rate
scrambling
chip
rate
chip
rate
Pulse
shaping
filter
Pulse
shaping
filter
Modulation
(BPSK)
Channel
h(1)(t)
Channel
h(2)(t)
MAI
dK(i)
( user K)
Modulation
(BPSK)
10 January,2002
Pulse
shaping
filter
Seminar of Master Thesis
Channel
h(K)(t)
17
Helsinki University of Technology
Department of Electrical and Communication Engineering
6. Simulation (2)
• Simulation block diagram of
2- D RAKE receiver in uplink WCDMA
Spatial processing
●

●
●
●
●
T
o

●
●

W1,M
●
●
User 1
Temporal processing (RAKE)
* (t-τ )
1
M
W1,2

T
o
●
* (t-τ )
1
2
W1,1
●
()

T
o

()
d 1 (i )
()
*
c~ 1(t-τ1)
●
Multiuser

n(t)
User K
●
●
●
●
●
●

WK,M
●
●

o

WK,1
()

Detection
* (t-τ )
1
M
WK,2

T
o
●
T
* (t-τ )
1
2

T
o
()

d K (i )
()
*
c~ K(t-τ1)
10 January,2002
Seminar of Master Thesis
18
Helsinki University of Technology
Department of Electrical and Communication Engineering
6. Simulation (3)
•
Channel: ray tracing channel model
The simulation area : the campus area of Dresden University of Technology.
Simulation area
MS location
BTS
10 January,2002
Seminar of Master Thesis
19
Helsinki University of Technology
Department of Electrical and Communication Engineering
6. Simulation Results(4)
• Assume all the users randomly access the channel, and the PN code of each user is acquired
and synchronized perfectly in the base station.
• Assume the number of fingers in RAKE receiver equal to the number of multipath
components.
• The channel parameters are updated symbol by symbol, i.e. channel varies with time.
The system performance of 1D RAKE and conventional
matched filter receiver for
single user
System performance of 1- D RAKE
receiver with Decorrelating Detector and
linear MMSE
DD: Decorrelating Detector
LMMSE
SA: smart antenna for spatial processing
PG: Processing Gain
MUD: Multiuser Detection
10 January,2002
Seminar of Master Thesis
20
Helsinki University of Technology
Department of Electrical and Communication Engineering
6. Simulation Results(5)
The system performance of 1D RAKE with different
Processing Gain
The system performance of 1-D RAKE with
spreading by Walsh codes and scrambling
by Gold codes
spreading and scrambling by random codes
The system performance of 2-RAKE receiver with RLS algorithm with adaptive
memory for spatial processing and MUD
System performance
0
2 antenna elements
3 active users
10
Conventional
Rake
SA RAKE
SA RAKE adaptive MUD(RLS)
-1
10
BER
3 antenna elements
3 active users
-2
10
0
10 January,2002
10
20
30
40
SNR (dB)
Seminar of Master Thesis
50
60
70
80
DD: Decorrelating Detector
LMMSE
SA: smart antenna for spatial processing
PG: Processing Gain
MUD: Multiuser Detection
21
Helsinki University of Technology
Department of Electrical and Communication Engineering
7. Conclusion
• The simulation results have shown that spreading,
RAKE receiver, multiuser detection and smart
antenna are very important techniques to improve
WCDMA system performance and increase system
capacity.
10 January,2002
Seminar of Master Thesis
22