Design of adaptive MIMO system using linear

dispersion code
Mabruk Gheryani, Zhiyuan Wu, and Yousef R. Shayan
Concordia University, Department of Electrical Engineering
Montreal, Quebec, Canada
email: (m gherya, zy wu, yshayan)@ece.concordia.ca
Abstract In this paper, we develop a new design for selectionmode adaptation using linear dispersion code. A new adaptive
parameter, called space-time symbol rate, can be applied due to
the use of the linear dispersion code. An adaptive algorithm for
the selection-mode adaptation is proposed. Based on the proposed
algorithm, two adaptive techniques using constellation and spacetime symbol rate are studied, respectively. If constellation and
space-time symbol rate are considered jointly, more selection
modes can be available. Theoretical analysis demonstrates that
the average transmission rate of selection-mode adaptation can
be improved in this case. Simulation results are provided to show
the benefits of our new design.

I. I NTRODUCTION
The demand for bandwidth efficiency in wireless communications has experienced an unprecedented growth. One significant advancement to improve radio spectrum efficiency is the
use of multiple-input-multiple-output (MIMO) technology [1]
[2]. Space-time (ST) codes are the most promising technique
for MIMO systems [3] [4]. Due to battery life and device
size, the power available for radio communications is limited.
Under this power constraint, adaptive technique can cooperate
with MIMO technology to further exploit radio spectrum [5]
[6].
In an adaptive system, a feedback channel is utilized to
provide channel state information (CSI) from the receiver
to the transmitter. According to the feedback of CSI, the
transmitter will adjust transmission parameters, such as power
allocation, modulation, coding rate, etc. This is conditioned
by the fact that the channel stays relatively constant before
the transmitter receives the CSI and then transmits next data
block accordingly. That is, the channel is slow. Many of
adaptive MIMO schemes have been proposed, such as waterfilling-based schemes [1] [7]- [9] and various beamforming
schemes [6] [10]- [13]. The above schemes often need nearperfect CSI feedback for adaptation calculation and consume
large feedback bandwidth. In practice, the channel estimation
will exhibit some inaccuracy depending on the estimation
method. The receiver will need time to process the channel
estimate and the feedback is subject to some transmission
delay. The transmitter needs some time to choose a proper
code, and there are possible errors in the feedback channel.
All these factors make the CSI at the transmitter inaccurate.
Additionally, the feedback bandwidth is often limited. In these
cases, adaptive schemes with a set of discrete transmission

modes are often more preferable. We can call them selectionmode adaptation. At the receiver, the channel is measured and
then one transmission mode with the highest transmission rate
is chosen, which meanwhile meets the BER requirement. The
optimal mode is fed back to the transmitter.
For selection-mode MIMO adaptation, the most convenient
adaptive parameter is constellation size for uncoded systems.
For example, constellation adaptation, such as M-QAM, is
applied to space-time block code (STBC) [14] and to spacetime trellis code (STTC) [15]. The disadvantage of these
schemes, is that they are not flexible for different rates, which
is the key requirement in the future wireless communications.
Additionally, the gap between the available transmission rates
are often very large due to the use of discrete constellations
[11].
In this study, we propose to apply linear dispersion code
(LDC) for adaptation. This is because it subsumes many existing block codes as its special cases which allows suboptimal
linear receivers with greatly reduced complexity, and provides
flexible rate-versus-performance tradeoff [16] [17]. The LDC
breaks the data stream into sub-streams that are dispersed over
space and time and then combined linearly at the transmitter
[16]. Particularly, since the LDC is applied, it makes ST
symbol rate available for adaptation. By adjusting this new
parameter together with constellation size, more available
transmission modes can be provided. Hence, the throughput
under a power constraint can be further improved while the
target bit error rate (BER) is satisfied.
II. S YSTEM M ODEL
In this study, during one ST modulation block, the channel
is assumed to be the same as estimated at the receiver.
Furthermore, the channel is assumed to be a Rayleigh flat
fading channel with Nt transmit and Nr receive antennas.
Lets denote the complex gain from transmit antenna n to
receiver antenna m by hmn and collect them to form an
Nr Nt channel matrix H = [hmn ], known perfectly to the
receiver. The entries in H are assumed to be independently
identically distributed (i.i.d.) symmetrical complex Gaussian
random variables with zero mean and unit variance.
The selection-mode adaptive system is depicted in Fig. 1. In
this system, the information bits are first mapped into symbols.
After that, the symbol stream is parsed into blocks of length
L. The symbol vector associated with one modulation block is

Binary
Info.
source

and variance z2 , and P is the average energy per channel

use at each receive antenna. Since the LDC is linear, an
MMSE detector can be applied as suboptimal receiver due
to its simplicity and good performance [21].

Binary
Info.
Out

Ant-1
1

Ant-1

M1
Nt
Constellation
Mapper

ML
Or
MMSE

S/P

De-Mapper

Ant-Nr
Ant-Nt

2Q

ML
Nt

L /T
Feedback
Selection
Mode

Fig. 1.

Selection-mode adaptive system block diagram.

denoted by x = [x1 , x2 , . . . , xL ]T with xi {m |m =

0, 1, . . . , 2Q 1, Q 1}, i.e., a complex constellation of size
2Q , such as 2Q -QAM). The average symbol energy is assumed
2Q
1
P
to be 1, i.e., 21Q
|m |2 = 1. Each block of symbols will
m=0

be mapped by the ST modulator to a dispersion matrix of size

Nt T and then transmitted over the Nt transmit antennas
over T channel uses. The following model will be considered
in this study, i.e.,
L
X
X=
Mi xi
(1)

III. D ESIGN OF S ELECTION -M ODE A DAPTATION

The general idea of selection-mode adaptation is to maximize the average transmission rate by choosing a proper transmission mode from a set of available modes. Based on some
certain strategy, the transmitter is informed by the receiver to
increase or decrease the transmission rate depending on the
channel condition, i.e., CSI. For selection-mode adaptation,
the signal-to-noise ratio (SNR) will be considered as a proper
metric. The corresponding adaptive algorithm is proposed as
follows.
1) Find the SNR, saying o , at the receiver;
2) Find the BERs of each mode at the obtained SNR o
from BER curves by experiment;
3) Select a proper transmission mode with the maximum
rate while satisfying the target BER;
4) Feed back the selected mode to the transmitter.
We can formulate the selection of transmission modes as
follows.
opt = arg
max
Rn
(7)
{n ,n=1,2,...,N }

i=1

where Mi is defined by its L Nt T dispersion matrices

Mi = [mi1 , mi2 , . . . , miT ]. The so-obtained results can be
extended to the model in [16]. With a constellation of size 2Q ,
the data rate of the space-time modulator in bits per channel
use is
Rm = Q L/T
(2)
Hence, one can adjust ST symbol rate L/T and constellation
size Q according to the feedback from the receiver.
In our design example, the dispersion matrices are given by
M(k1)Nt +i = diag[fk ]P(i1)

(3)

for k = 1, 2, . . . , Nt and i = 1, 2, . . . , Nt , P is the permutation

matrix of size Nt and given by

01(Nt 1)
1
P=
(4)
INt 1
0(Nt 1)1
where fk denotes the k-th column vector of F. F = [fmn ] is
a Fast Fourier Transform (FFT) matrix and fmn is calculated
by
1
fmn = exp(2j(m 1)(n 1)/Nt )
(5)
Nt
At the receiver, the received signals associated with one
modulation block can be written as
r
r
L
X
P
P
HX + Z =
H
Mi xi + Z
(6)
Y=
Nt
Nt i=1
where Y is a complex matrix of size N r T whose (m, n)th entry is the received signal at receive antenna m and time
instant n, Z is the additive white Gaussian noise matrix with
i.i.d. symmetrical complex Gaussian elements of zero mean

subject to
BERn (o ) BERtarget

(8)

where {n , n = 1, 2, . . . , N } is the set of transmission

modes, Rn is the rate of transmission mode n , BERn (o )
is the BER of transmission mode n at SNR o and
BERtarget is the target BER. Without loss of generality, we
assume R1 < R2 < . . . < RN . opt is the optimal
transmission mode at SNR o .
Below, we consider the average transmission rate using the
proposed adaptive algorithm. Let n denote the minimum
SNR satisfying the following condition.
n = arg min[BERn () BERtarget )]

(9)

That is, for the SNR region n n+1 , the transmission rate Rn (i.e., the transmission mode n ) should be
selected while the target BER is satisfied.
Then, the average transmission rate is
Z
N
X
n+1
=
R
Rn
p ()d
(10)
n=1

where p () is the probability density function (PDF) of

the SNR and N +1 = . Maximization of the average
can be solved using Lagrange multitransmission rate R
pliers. However, due to the structure of both the objective
function and the inequality constraint, an analytical solution
is extremely difficult to find. Therefore, we will find the
SNR region corresponding to each transmission mode by
measurement.
In our simulations, we assume Nt = Nr = 4. First, we
perform constellation adaptation alone with a fixed ST symbol

10

TABLE I
A DAPTIVE CONSTELLATION WITH ST SYMBOL RATE L/T = 1, 2, 3, 4

10

8PSK1layer
QPSK1layer
BPSK1layer
16QAM1lyaer

BPSK2layer
QPSK2layer
8PSK2layer
16QAM2layer

10

10

BER

BER

10

10

10

10
4

10

6
SNR(dB)

10

12

14

10

16

(a) L/T = 1

8
10
SNR(dB)

12

14

16

18

20

(b) L/T = 2

10

BPSK3Layer
QPSK3layer
8PSK3layer
16QAM3layer

10

BPSK4layer
QPSK4layer
8PSK4layer
16QAM4layer

10

10

BER

BER

10

10

10

10

10

10

15

20

25

30

10

15

SNR(dB)

20

25

30

35

SNR(dB)

(c) L/T = 3
Fig. 2.

MODE
0
1
2
3
4
MODE
0
1
2
3
4
MODE
0
1
2
3
4
MODE
0
1
2
3
4

Constellation
BPSK
QPSK
8PSK
16QAM
Constellation
BPSK
QPSK
8PSK
16QAM
Constellation
BPSK
QPSK
8PSK
16QAM
Constellation
BPSK
QPSK
8PSK
16QAM

L/T
1
1
1
1
L/T
2
2
2
2
L/T
3
3
3
3
L/T
4
4
4
4

Rm
1
2
3
4
Rm
2
4
6
8
Rm
3
6
9
12
Rm
4
8
12
16

T
Q
< 0.6309
0.6309 11 < 0.1893
0.1893 21 < 3.384
3.384 31 < 11.7479
41 11.7479
2
Q
< 0.8385
0.8385 12 < 1.4058
1.4058 22 < 5.3886
5.3886 32 < 15.4452
42 15.4452
3
Q
< 3.1014
3.1014 13 < 4.4833
4.4833 23 < 8.9696
8.9696 33 < 26.5898
43 26.5898
4
Q
< 8.1509
8.1509 14 < 14.2812
14.2812 24 < 24.2533
24.2533 34 < 30.8208
44 30.8208

(d) L/T = 4
Adaptive Constellation.

rate. Secondly, we perform the ST symbol rate adaptation

alone with a fixed constellation. Finally, we will consider these
two parameter jointly to maximize the average transmission
rate meanwhile maintaining the target BER, which is equal to
103 in our design examples.
A. Adaptation Using Variable Constellations
Although the system design for continuous-rate scenario
provide intuitive and useful guidelines [11], the associated
constellation mapper requires high implementation complexity. In practice, discrete constellations are preferable. That is,
Q only takes integer number, such as Q = 1, 2, 3, ..... For
a given adaptive system, we can adjust the constellation to
maximize the transmission rate meanwhile keeping the target
BER satisfied. The proposed adaptive algorithm is applied
to the case. Here, we only consider BPSK (Q = 1), QPSK
(Q = 2), 8PSK (Q = 3) and 16QAM (Q = 4) as examples.
That is, n {BP SK, QP SK, 8P SK, 16QAM } with a
fixed ST symbol rate. The optimal transmission mode is
selected by the proposed adaptive algorithm, i.e., by equation
(7) and (8). Simulation results are shown in Fig. 2, where
each subfigure has its own ST symbol rate. We summarize
L
our simulation results in Table I. In the following context, QT
denotes the SNR associated with the transmission mode with
2Q constellation and TL ST symbol rate.
B. Adaptation Using Variable ST Symbol Rate
In other existing schemes, only the orthogonal designs,
such as Alamouti scheme, are applied as the ST modulation.

In this case, the most convenient adaptive parameter is the

constellation size. For our adaptive scheme, the application
of LDC makes another adaptive parameter available, i.e., ST
symbol rate. In this subsection, we fix the constellation size
but adjust the ST symbol rate for adaptation. Additionally, one
advantage of using ST symbol rate is that it is easier to change
ST symbol rate than constellation size for adaptation as can
be seen in Fig. 1. The proposed adaptive algorithm described
by (7) and (8) can be applied to ST symbol rate adaptation.
Note that, this system with 4 transmit antennas can have
16 choices of ST symbol rates, i.e., ( 14 16
4 ). For
convenience and less complexity, we use 4 choices, i.e., TL =
1, 2, 3, 4. That is, n { TL = 1, TL = 2, TL = 3, TL = 4} with
a fixed constellation. In the following context, the integer of
L
T is referred as layer. The simulation results are shown in
Fig. 3, where each subfigure has its own constellation. We
summarize these results in Table II.
IV. J OINT A DAPTIVE T ECHNIQUE
As shown in the previous two subsections, either constellation adaptation or ST symbol rate adaptation can increase the
average transmission rate while the given BER is satisfied as
compared to non-adaptive schemes. However, we can further
improve the average transmission rate by applying a joint
adaptation. The joint adaptation is performed by choosing
the best pair of constellation size and ST symbol rate. The
available transmission modes are increased. That is,
n {(BP SK, TL = 1),
(QP SK, TL = 1),
(8P SK, TL = 1),
(16QAM, TL = 1),

...,
...,
...,
...,

(BP SK, TL = 4),

(QP SK, TL = 4),
(8P SK, TL = 4),
(16QAM, TL = 4)}

TABLE III
J OINT A DAPTATION OF ST SYMBOL RATE AND CONSTELLATION SIZE
BER BPSK 4X4

BER 4PSK 4X4

10

10

MODE
0
1
2
3
4
5
6
7

2layer
3layer
4layer
1layer

3Layer
4layer
1layer
2layer
1

10
1

10

10
2

BER

BER

10

10

10

10

10

10

Constellation
BPSK
QPSK
QPSK
QPSK
8PSK
8PSK
16QAM

L/T
1
1
2
3
3
4
4

Rm
1
2
4
6
9
12
16

T
Q
< 0.6309
0.6309 11 < 0.1893
0.1893 12 < 1.4058
1.4058 22 < 4.4833
4.4833 32 < 8.9696
8.9696 33 < 24.2533
24.2533 43 < 30.8208
44 30.8208

4
SNR(dB)

10

10

(a) BPSK (Q = 1)

8
10
SNR(dB)

12

14

16

18

BER 16QAM 4X4

10

10
1layer
2layer
3layer
4layer

3layer
1lyaer
2layer
4layer

10

BER

BER

10

10

10

10

10

10

20

(b) QPSK (Q = 2)

BER 8PSK 4x4

10

15
SNR(dB)

20

10

25

10

15

20

25

30

35

SNR(dB)

(c) 8PSK (Q = 3)
Fig. 3.

(d) 16QAM (Q = 4)
Adaptive ST symbol rate.

TABLE II
A DAPTIVE ST SYMBOL RATE WHEN CONSTELLATION IS BPSK, QPSK,
8PSK AND 16QAM, RESPECTIVELY.
MODE
0
1
2
3
4
MODE
0
1
2
3
4
MODE
0
1
2
3
4
MODE
0
1
2
3
4

Constellation
BPSK
BPSK
BPSK
BPSK
Constellation
QPSK
QPSK
QPSK
QPSK
Constellation
8PSK
8PSK
8PSK
8PSK
Constellation
16QAM
16QAM
16QAM
16QAM

L/T
1
2
3
4
L/T
1
2
3
4
L/T
1
2
3
4
L/T
1
2
3
4

Rm
1
2
3
4
Rm
2
4
6
8
Rm
3
6
9
12
Rm
4
8
12
16

T
Q
< 0.6309
0.6309 11 < 0.8385
0.8385 12 < 3.1014
3.1014 13 < 8.1509
14 8.1509
2i
< 0.1893
0.1893 21 < 1.4058
1.4058 22 < 4.4833
4.4833 23 < 14.2812
24 14.2812
3i
< 3.384
3.384 31 < 5.3886
5.3886 32 < 8.9696
8.9696 33 < 24.2533
34 24.2533
4i
< 11.7479
11.7479 41 < 15.4452
15.4452 42 < 26.5898
26.5898 43 < 30.8208
44 30.8208

We can reduce the gap between the selection modes further

by adding more choices of the transmission rates. For the
target BER, a scheme with the joint adaptation can improve
the average transmission rate significantly as compared to the
two techniques in the previous subsections.
We conclude the result in Table III.
From the simulation results, we have the following observations:
If the ST symbol rate is reduced, the slope of the
associated BER curve becomes steeper, which suggests
a larger diversity;
If the constellation size is reduced, the BER curve will
shift to left with the similar slope, which suggests the
diversity keeps the same but the coding gain is improved.
There exists a tradeoff between diversity gain and multiplexing
gain [22]. However, this tradeoff can not provide insight for
the adaptive system with discrete constellations. From the
above observations, we find that we can improve data rate
by using the two adaptive parameters jointly. Specifically, in
some cases, we can adjust constellation size to improve rate
and performance; which in the other cases, we will adjust ST
symbol rate, i.e., multiplexing gain, for adaptation. To proceed,
we have the following proposition.
Proposition 1: The average transmission rate in the adaptive selection-mode system can be improved by adding more
possible transmission modes providing higher data rate than
the corresponding original mode at the same SNR region.
Proof: Let us define the SNR regions of our adaptive system
using one set of selection modes as follows.
<1 1 < < 2

associated

with R1

..
.
<i i < < i+1

associated with Ri

If we add more possible selection modes, the SNR regions

will be changed as follows.
0

<1 1 < < 1

0

<1 1 < < 2

associated

with R1

associated

with R1

..
.
0

<i i < < i

associated with Ri

20

18

R EFERENCES

ASE Joint Adaption (Q,L/T)

ASE changing constellation size(BPSK,QPSK,8PSK,16QAM with L/T=4
ASE changing ST symbol rate(L/T=1,2,3,4) with 16QAM

Average Spectral Efficiency bps/channel use

16

14

12

10

0
5

10

15
20
SNR[dB]

25

30

35

40

Fig. 4. Average spectral efficiency comparison for the three adaptive schemes.

<i i < < i+1

associated with Ri

We assume Ri > Ri for any i. The total average rate for

original scheme can be written as
X Z i+1
=
R
Ri
p ()d
i

The total average rate when for the scheme with more
transmission modes can be written as
Z i0
Z i+1
X
0

A=
(Ri
p ()d + Ri
p ()d)
i

It is obvious that

A > R

In Fig. 4, we compare the average spectral efficiency (ASE)

for the three adaptive techniques. As can be seen from Fig. 4,
the ASE of the joint adaptive scheme outperforms the other
two schemes significantly from 0dB to 25dB. At high SNR
(larger than 25 dB), three schemes have the same performance.
As predicted by Proposition 1, if there are more available
modes, the ASE can be improved further.
V. C ONCLUSIONS
In this paper, we proposed a novel adaptive design with
discrete selection modes, in which the linear dispersion code
is applied. Since the high-rate LDC is applied, it makes spacetime symbol rate available for adaptation. An adaptive algorithm is proposed for selection-mode adaptation. Based on the
proposed algorithm, two adaptation techniques using constellation and space-time symbol rate are studied, respectively. With
joint adaptation of space-time symbol rate and constellation
size, more transmission modes can be provided to reduce rate
gap among transmission modes. Theoretical analysis shows
that the average transmission rate can be improved with
more available transmission modes. Additionally, with spacetime symbol rate of the linear dispersion code, the adaptive
design can be simplified and various levels of diversity and
multiplexing gain can be provided. Simulation results were
provided to demonstrate merits of the joint adaptation of
constellation and space-time symbol rate.

[1] I. E. Telatar, Capacity of multi-antenna Gaussian channels, Eur. Trans.

Telecom., vol 10, pp. 585-595, Nov. 1999.
[2] G. J. Foschini, M. J. Gans, On limits of wireless communications in
a fading environment when using multiple antennas, Wireless Personal
Communications, vol. 6, no. 3, pp. 311-335, 1998.
[3] S. Alamouti, A simple transmitter diversity scheme for wireless communications, IEEE J. Select. Areas Commun., vol. 16, pp. 1451-1458, Oct.
1998.
[4] V. Tarokh, N. Seshadri, and A. Calderbank, Space-time codes for
high data rate wireless communications: Performance criterion and code
construction, IEEE Trans. Inform. Theory, vol. 44, pp. 744-765, Mar.
1998.
[5] D. Gesbert, R. W. Heath,and S. Catreux,V. Erceg Adaptive modulation
and MIMO coding for broadband wireless data networks, in 2002 IEEE
Communications Magazine vol. 40, pp. 108-115 , June 2002.
[6] S. Zhou and G. B. Giannakis, Optimal transmitter eigen-beamforming
and space-time block coding based on channel mean feedback IEEE
Transactions on Signal Processing, vol. 50, no. 10, October 2002.
[7] Z. Luo, H. Gao,and Y. Liu,J. Gao Capacity limits of time-varying MIMO
channels, IEEE International Conference On Communications vol.2,
Mar. 2003.
[8] Z. Zhou and B. Vucetic Design of adaptive modulation using imperfect
CSI in MIMO systems, 2004 Eelectronics Letters vol. 40 no. 17 August
2004.
[9] X. Zhang and B. Ottersten, Power allocation and bit loading for spatial
multiplexing in MIMO systems, IEEE Int. Conf.on Acoustics, Speech,
and Signal Processing, 2003. Proceedings (ICASSP 03) vol.5 pp. 54-56,
Apr. 2003.
[10] J. K. Cavers, Single-user and multiuser adaptive maximal ratio transmission for Rayleigh channels, IEEE Trans. Veh. Technol., vol. 49, no.
6, pp. 20432050, Nov. 2000.
[11] P. Xia,and G. B. Giannakis, Multiantenna adaptive modulation with
beamforming based on bandwidth-constrained feedback IEEE Transactions on Communications, vol.53, no.3, March 2005.
[12] Bishwarup Mondal and Robert W. Heath, Jr., Performance analysis
of quantized beamforming MIMO systems IEEE Trans. on Signal
Processing , vol. 54, no. 12, DECEMBER 2006.
[13] S. Zhou and G. B. Giannakis, How accurate channel prediction
needs to be for transmit-beamforming with adaptive modulation over
Rayleigh MIMO channels, IEEE Trans. Wireless Comm., vol.3, no.4,
pp. 12851294, July 2004.
[14] Youngwook KO and Cihan Tepedelenlioglu, Space-time block coded
rate-adaptive modulation with uncertain SNR feedback IEEE Signals,
Systems and Computers, vol.1, pp 1032- 1036, Nov. 2003.
[15] T. H. Liew, B. L. Yeap, C. H. Wong, and L. Hanzo, Turbo-coded
adaptive modulation versus spacetime trellis codes for transmission over
dispersive channels, IEEE Trans. om Wireless Comm., vol.3, no. 6, pp
2019-2029, Nov. 2004.
[16] B. Hassibi and B. Hochwald, High-rate codes that are linear in space
and time, IEEE Trans. Inform. Theory, vol. 48, pp. 1804-1824, July
2002.
[17] Z. Wu and X. F. Wang, Design of coded space-time modulation,
IEEE International Conference on Wireless Networks, Communications
and Mobile Computing, vol. 2, pp. 1059-1064, Jun. 13-16, 2005.
[18] R. W. Heath and A. Paulraj, Linear dispersion codes for MIMO systems
based on frame theory, IEEE Trans. on Signal Processing, vol. 50, No.
10, pp. 2429-2441, October 2002.
[19] G. J. Foschini, Layered space-time architecture for wireless communication in fading environments when using multiple antennas, Bell labs.
Tech. J.,vol. 1, no. 2, pp. 41-59, 1996.
[20] H. El Gamal and A. R. Hammons Jr., A new approach to layered spacetime coding and signal processing, IEEE Trans. Inf. Theory, vol. 47, pp.
2321-2334. Sep. 2001.
[21] R. Lupas and S. Verdu, Linear multiuser detectors for synchronous
code-division multiple-access channels, IEEE Trans. inform. Theory, vol.
35, pp. 123-136, Jan. 1989.
[22] L. Zheng and D. Tse, Diversity and multiplexing: A fundamental
tradeoff in multiple antenna channels IEEE Trans. Inform. Theory, vol.
49, pp. 1073-96, May 2003.