Transcript Joey Han

Adaptive radio-frequency resource
management for MIMO MC-CDMA on
antenna selection
Jingxu Han and Mqhele E Dlodlo
Department of Electrical Engineering
University of Cape Town
What is MIMO?
 MIMO (Multiple Input Multiple Output) is a new
wireless technology conceived in the mid 90’s
 Greatly improves the reliability, range and
robustness of existing wireless communication
systems
 Can provide high data rate (channel capacity) high
QoS (low bit error rate)
 Selected for many standards, such as W-CDMA
UCT-COE Seminar
000 Page 2 April 8, 2016
Introduction of MC CDMA
 Multi-carriers CDMA (MC CDMA) is a multiple
access technology used in wireless communication
systems based on OFDM
 a strong candidate as an access technique for
downlink of broadband cellular systems
 Obtains benefits from both MC and CDMA
techniques: high spectral efficiency, robustness in
frequency selective channels
UCT-COE Seminar
000 Page 3 April 8, 2016
Combination of MIMO MC-CDMA systems
 A promising technology for 4G wireless
communication since:
1 high date rate
2 high Qos
3 multiple users communication
4 high spectral efficiency
UCT-COE Seminar
000 Page 4 April 8, 2016
Why antenna selection?
 An inherent drawback for multiple antenna system
is the increased hardware complexity due to the
requirement for radio-frequency chain.
 Corresponding to power amplifiers, mixers analog
to digital converter ect.
 An ideal approach to reduce the complexity of a
multiple antennas system while maintaining the
advantages
UCT-COE Seminar
000 Page 5 April 8, 2016
Ideas behind antenna selection
 Choose a subset of best performance antennas
either in transmission and/ or reception to feed to
the RF chain for signal processing
 Existing algorithms:
 1) maximum signal to noise ratio
 2) maximum channel mean
 3) minimum instantaneous probability of error
UCT-COE Seminar
000 Page 6 April 8, 2016
Proposing algorithm
 An improved low complexity algorithm proposed
based on maximum ratio method
 the method is simplified to choose the best
performance antenna from all the possible
antennas, then choose the best performance
antenna from the rest possible antennas instead of
calculate Frobenius norm for all the possible
antennas
UCT-COE Seminar
000 Page 7 April 8, 2016
Complexity comparison
 For instance: selecting 2 antennas out of 10
antennas
 Old algorithm calculation complexity is 45
 Correspondingly the novel one is 19
UCT-COE Seminar
000 Page 8 April 8, 2016
Adaptive RF management
 Aim to flexibly manage RF resource for MIMO
MC-CDMA systems based on antenna selection
 1) Set target bit error rate
 2) find out minimum number of antennas to feed RF need
 3) adaptive increasing the number of antennas on
increment of users or interferences
UCT-COE Seminar
000 Page 9 April 8, 2016
Simulation results
performance of a 2by1 MIMO MC-CDMA system on antenna selection
0
10
-1
10
-2
SER
10
-3
10
32 users
32 users
16 users
16 users
1 user
1 user
-4
10
-5
10
0
1
2
3
4
5
SNR
6
7
8
9
10
UCT-COE Seminar
000 Page 10 April 8, 2016
performance of different selected number of transmitters
-1
10
2
3
4
5
6
7
-2
BER
10
out
out
out
out
out
out
7
7
7
7
7
7
-3
10
-4
10
-5
10
0
1
2
3
4
5
SNR
6
7
8
9
10
UCT-COE Seminar
000 Page 11 April 8, 2016
Impact on various number of users
-1
10
-2
BER
10
-3
10
2 out
3 out
4 out
5 out
6 out
7 out
-4
10
7
7
7
7
7
7
-5
10
0
5
10
15
20
the number of users
25
30
35
UCT-COE Seminar
000 Page 12 April 8, 2016
performance of the proposed adaptive system with target BER to 1%
-1
10
BER
adaptive
-2
10
-3
10
0
5
10
15
20
the number of users
25
30
35
UCT-COE Seminar
000 Page 13 April 8, 2016
performance of the proposed adaptive system with target BER to 0.1%
-2
10
BER
adaptive
-3
10
-4
10
0
2
4
6
SNR
8
10
12
UCT-COE Seminar
000 Page 14 April 8, 2016
Conclusion
 A novel transmission adaptive technique for MIMO
MC-CDMA system as an aim to flexibly manage
radio frequency resource based on antenna
selection is presented
 The simulation results, depending on both numbers
of users and signal to noise ratio (SNR) value,
show significant gains compared to fixed systems
for a predefined target bit error rate (BER)
UCT-COE Seminar
000 Page 15 April 8, 2016
Future work
 Performance investigation for the real MIMO
channels are needed
 Effective feedback path is necessary to insure the
system, and performance over fast fading channel
is also an interesting topic.
UCT-COE Seminar
000 Page 16 April 8, 2016
Thanks
Questions?