Layered Space-Time Architecture For Wireless Communication in A Fading Environment When Using Multi-Element Antennas
Layered Space-Time Architecture For Wireless Communication in A Fading Environment When Using Multi-Element Antennas
Introduction
This paper describes a new point-to-point communication architecture employing an equal number
of antenna array elements at both the transmitter and
receiver. The architecture is designed for a Rayleigh
fading environment in circumstances in which the
transmitter does not have knowledge of the channel
characteristic. This new communication structure,
termed the layered space-time architecture, targets application in future generations of fixed wireless systems,
bringing high bit rates to the office and home. The
architecture might also be used in future wireless local
area network (LAN) applications for which it promises
extraordinarily high bit rates.
The architecture is a method of presenting and
41
42
r=
/ nT h s + .
(1a)
+ 1 Normal 0,1 /
2 .
Capacity
The viewpoint assumed in treating capacity is discussed next. We stress that capacity is a limit to errorfree bit rate that is provided by information theory,
Transmitter
Receiver
(9, 12)
ssor
Proce
Figure 1.
Shorthand notation (nT , nR ) for the number of transmit antennas nT and receive antennas nR (dipole antennas shown).
Long-Burst Perspective
Communication in long bursts means bursts having many symbolsso many that an infinite time
horizon information-theoretic description of communication portrays a meaningful idealization. Yet bursts
are assumed to be of short enough duration that a
channel is essentially unchanged during a burst. The
channel is assumed to be unknown to the transmitter
but learned (tracked) by the receiver. The channel
might change considerably from one burst to the next.
By a channel being unknown to the transmitter,
we mean that the realization of H during a burst is
unknown. Actually, the average SNR value and even
n R might not be known to the transmitter.
Nonetheless, for purposes of this discussion, these two
parameters are considered to be known. The reason
for this is that at the transmitter, one assumes that
communication is taking place with a user for which
at least a certain nR and average SNR are available.
These minimum values represent what the transmitter conservatively uses to determine a capacity value
43
Section of available
surface of a base
or laptop
Dielectric
surface
/2
/2
/4
Metal
surface
/2
Dielectric
layer
/2
Figure 2.
Section of casing paved with half-wavelength lattice.
44
C>
log
k= 1
1 + ( / n) 2k b/s /Hz .
(3)
300
10
24 dB
0 dB
200
Capacity (b/s/Hz)
8
18 dB
150
12 dB
100
0 dB
0
0
10
20
30
(a)
40
50
6
4
12 dB
18 dB
24 dB
60
10
20
24 dB
6
18 dB
12 dB
6 dB
0 dB
20
30
50
60
0.25
6 dB
0.20
0.15
0.10
12 dB
0.05
24 dB
0
10
40
0.30
30
(c)
10
6 dB
6 dB
50
Capacity (b/s/Hz/dimension)
Capacity (b/s/Hz)
250
Capacity (b/s/Hz/dimension)
40
50
60
10
20
18 dB
30
40
50
60
(d)
(b)
Figure 3(a).
Capacity in b/s/Hz versus the number of antenna elements at each site.
Figure 3(b).
Capacity in b/s/Hz/dimension versus the number of antenna elements at each site.
Figure 3(c).
Capacity in b/s/Hz versus the number of antenna elements at each site.
Figure 3(d).
Capacity in b/s/Hz/dimension versus the number of antenna elements at each site.
2
usedfor example, 2k to denote directly a chi-
45
Primitive
data stream
DEMUX
Equal
rates
Layer 1
(mod/code)
Layer 2
(mod/code)
Layer n
(mod/code)
Antennas
Figure 4.
Transmission process using space-time layering.
C = log2 1 +
22n
b/s/Hz .
(4)
46
log (1 + x) dx
0
= 1 + 1 log 2 (1 + ) log 2 e.
(5)
Time
4
3
Time
2
1
7
Figure 5.
Flow of nominal processing time for a received signal.
Mathematical Background
The following linear algebra helps clarify the
architecture. Let H j with 1 j n denote the n
columns of the H matrix ordered left to right so
that H = H1, H2, ...Hn . For each k such that
47
Space
(associated
transmitter
element)
Detect
now
6
5
4
3
2
1
f
e
d
c
b
a
0
Detect
later
a
f
e
d
b
a
f
e
c
b
a
f
Canceled
d
e
f
Nulled
c
d
e
b
c
d
a
b
c
a
b
a
5
a
f
e
d
c
b
7
b
a
f
e
d
c
8
c
b
a
f
e
d
9
10
Time
Reconstruct signals for five underlying layers
and then subtract off interference from the
previously detected bits.
Avoid interference from five overlying layers
by forming decision statistics that avoid
interference from those overlying layers.
Demod a
waveform on:
t < 2
2 t < 3
3 t < 4
4 t < 5
5 t < 6
6 t < 7
Avoiding interference
from transmitter antenna:
6
5 and 6
4, 5, and 6
3, 4, 5, and 6
2, 3, 4, 5, and 6
Figure 6.
Space-time layering in reception.
Transmission
In a spectrally economical system, the layered
space-time architecture described here would be
employed in conjunction with an efficient 1-D code.
The form of the code employed in a specific instance of
the architecture is not within the scope of this paper.
For expositional simplicity, however, it is best to begin
the description by considering some nonspecific block
code rather than a convolutional code implementation.
Figure 4 illustrates the transmission process. A
48
Is
layer index
equal to
final?
No
Increment
layer index
Yes
End
Figure 7.
Temporal view of the processing of successive
space-time layers.
49
Linear
combination
avoiding no
interference
Linear
combination
avoiding
interference
from transmit
antenna n
Linear
combination
avoiding
interference
from transmit
antennas n, n1 2
Sum of
n1 unavoided
interferences
+
Sum of
n2 unavoided
interferences
+
Figure 8.
Spatial view of receiver processing.
50
that need not be nulled are those that will be subtracted out. Of course, when nulling interferers, any
possible enhancement of the noise caused by the interference nulling process must be carefully assessed. As
explained later in reference to the mathematical setup
that has been carefully tailored for capacity analysis,
the noise assessment will be easy to do.
Figure 6 illustrates additional details of the steps
required for proceeding along iterated diagonal layers.
For expositional convenience, a repetitive abcdef labeling on the stack is included. Detection of the first complete diagonal a layer through which is drawn a
dashed diagonal line is described. Other layers, including boundary layers, are handled similarly. Boundary
layers are those layers involved with where a burst
starts or ends (those having fewer than six rectangles).
The first complete a layer comprises six parts,
aj(t) (j = 1, 2, ... 6), in which the subscript indicates at
Primitive
bit stream
n equal rate
substreams
1:n
DEMUX
Periodic t-varying
vector to avoid
interference from
detected bits*
Encoder
Periodic
t-varying
vector
Symbol
stream
Interference
vector from
periodically
varying set of
n1 antennas
AWGN
vector
+
n1 detected
substreams
<., .>
Compose vector
of interfering
detected symbols
Memory of
previously
detected symbols
Periodic t-varying
vector to avoid
interference from
undetected bits*
<., .>
Detected
bitstream
Interference-free
encoded stream
n:1
MUX
Decoder*
Figure 9.
System diagram of the processing involved at the receiver (discrete time baseband perspective).
what point in time the part that lasts for time units
begins. All layers relatively disposed to be located partially underneath this a layer are assumed already successfully detected while all layers disposed to be
partially above the a layer are yet to be detected. The
capacity associated with this case will also be found,
and then the capacity associated with the (n, n) case
will be apparent. (With block coding, a full layer could
correspond to exactly one block, although as pointed
out later, associating more than one block within a
layer can sometimes be advantageous).
Next, before computing capacity, a connection is
made to the earlier Mathematical Background subsection by pointing out the relevance of projecting a
51
Time
Space
No processing is done above and to the right
of the layer currently being processed. The
corresponding interferences do impair the
layer currently being processed.
6
5
4
3
2
Figure 10.
Blocks and sub-blocks in a space-time layer show how parallel processing can be used to advantage.
C = log2 1 + ( / 6) 12
b/s/Hz.
C = log2 1 + ( / 6) 10
b/s/Hz.
52
C = log2 1 + ( / 6) 2
b/s/Hz.
C = (1/6)
log
k =1
1+ ( / 6) 2k
2
b/s/Hz.
C=
log
k =1
1 + ( /6) 2k
2
b/s/Hz.
Code
Periodic
time-varying
channel
Decode
Code
Periodic
time-varying
channel
Decode
Periodic
time-varying
channel
Decode
Periodic
time-varying
channel
Decode
To
multiplexer
Code
To
sub-multiplexer
Code
Bit rate x/18
Subsystem at the top is replaced by three subsystems of one-third the bit rate.
Figure 11.
In parallel receiver processing, three time-varying channels run in parallel.
tem is given by
C=
k =1
log2 1 + ( / n) 2k b/s/Hz .
2
(6)
53
Robustness
In case the layered architecture described earlier
seems fragile, an explanation of why it can be quite
robust is included. At first, the architecture might seem
fragile. After all, the successful detection of each layer
relies on the successful detection of the underlying layers. Thus, any failure in any layer but the last will
likely cause the detection of all subsequent layers to
fail. A quantitative discussion is included in this subsection to illustrate that fragility generally is not a significant problem, especially when huge capacity is
available. As depicted in Figure 3(a), a huge capacity
can be a very reasonable assumption.
In practical implementations, the huge capacity
available can be invested in selecting a code that provides the required bit rate with very substantial error
protection. Let ERROR denote the event that a packet
(= long burst) contains at least one error for whatever
reason. Decomposing the ERROR event into two disjoint events gives
.
ERRORnonsupp denotes the event that channel realization simply does not support the required BER even
if receiver processing could be enhanced magically by
a genie removing interference entirely from all underlying layers.
ERRORsupp denotes the remaining ERROR events.
Assume that the required outage is 1%, packet size
(payload) is 10,000 bits, and a BER of 107 is required.
The extra capacity can be used instead to provide a
BER at least one order of magnitude lower. Because
10 4 10 8 = 10 4 , roughly one packet in 104 contains an error. Inflate the bit-error occurrences by
labeling all bits in such a packet in error. Such a
drastic inflation in the accounting of errors is a
harmless exaggeration. The reason for this is those
packets containing errors can be ascribed to outage
because they carry insignificant probability compared to Probability[ERRORnonsupp]. In effect, the
huge capacity available allows the luxury of taking
the perspective that ERROR = OUT. When the sys54
Related Options
This section discusses some modifications of the
communication architecture previously described.
Suggestions are provided as to what might be gained
or lost by these changes. Some of these items are preliminary ideas that are included as possibilities for
future research.
1
n
n 1
d on 0 4 n (0 elsewhere).
4
Lim n
Number of Channel Eigenvalues
n
~
(7)
1 1
d .
4
log (1 + 4x)
0
= log2 (1 + 4 )
ln 2
1
0
1 x
dx
x
x (1 x) + sin
1 + 4 x
(8)
dx .
(9a)
C given by (2).
The large asymptote is anticipated to be of interest in some applications. However, the following is
worth mentioning: One can derive that the advantage
of knowing the channel for large n but vanishingly
small is a factor of two in capacity
C ch _ knwn / n
/ l n 2 b/s/Hz /signal dimension
= 2C / n (as n , 0).
(9b)
55
stresses that three time-varying channels run in parallel. Even a sublayer could be decomposed into blocks
for the purpose of block coding, especially if n is large.
56
log
k =1
n
1+ ( / n) Hk Hk b/s/Hz
log2 1 + ( / n) 2n , k b/s/Hz .
2
(10)
k =1
(as 0)
Coding
We previously stressed block coding for expositional simplicity. In practice, convolutional codes or
a form of trellis-coded modulation for more bandwidth efficiency might have a role. (The block-convolutional distinction is blurred if a block code is
C=
n /2
log
k =1
57
No Cycling
As Figure 4 illustrates, cycling the substream to
antenna association was required at the transmitter. Is
this cycling really necessary? A straightforward but
tedious asymptotic argument shows that, in the limit
of large n, the receive diversity compensates for any
inferior Hijs. Consequently, the asymptotic linear
capacity growth with n also occurs even without
cycling.
58
Acknowledgments
Figures 1 and 2 were largely the creation of
M. J. Gans, who also advised the author on antenna
theory. Valuable discussions with I. Bar-David,
J. Mazo, A. Saleh, J. Salz, and L-F. Wei are also
gratefully acknowledged.
References
1. G. J. Foschini and M. J. Gans, On Limits of
Wireless Communication in a Fading
Environment When Using Multiple Antennas,
Wireless Personal Communications, accepted for
publication.
2. E. Csiszar and J. Korner, Information Theory:
Coding Theorems for Discrete Memoryless Systems,
Academic Press, New York, 1981.
3. S. Z. Stambler, Shannons Theorems for a
Complete Class of Discrete Channels Whose State
Is Known at the Output, Problems of Information
Transmission, No. 11, Plenum Press, New York,
November 1976, pp. 263-270.
4. S. Kullback, Information Theory and Statistics,
John Wiley and Sons, New York, 1959.
5. D. B. Osteyee and I. J. Good, Information Weight
of Evidence: the Singularity Between Probability
Measures and Signal Detection, Springer-Verlag,
New York, 1970.
6. M. S. Pinsker, Information and Information
Stability of Random Processes, Holden Bay,
San Francisco, 1964, Chapter 10.
7. R. Price, Nonlinearly Feedback-Equalized PAM
vs. Capacity for Noisy Filter Channels,
Proceedings of the IEEE International Conference on
Communications, June 1972, IEEE Publishing
59