Study of 2x2 Spatial Multiplexed System in 60 GHz

Indoor Environment
Savitha Manojna.D
Communications Engineering Research Group
Department of ECE
Amrita Vishwa Vidyapeetham
Coimbatore, India
sibylvista@gmail.com
Kirthiga.S, Jayakumar.M
Communications Engineering Research Group
Department of ECE
Amrita Vishwa Vidyapeetham
Coimbatore, India
s_krithiga@cb.amrita.edu


Abstract The Millimeter Waves are allocated with 60GHz
frequency range as an unlicensed band worldwide. Millimeter
waves become very useful for short-range communications
because of its high data rate, its large-available bandwidth of
7GHz and the Oxygen absorption present at that band. The
60GHz indoor channel contains much of multipath components
and needs the use of statistical parameters in modeling the
channel. At 60 GHz, the free space loss is higher and thus
requires the use of antennas with more pattern directivity along
with having small antenna dimensions. Considering these factors,
in this paper, Triple Saleh Valenzuela (TSV) model is chosen as a
suitable model for millimeter waves. The performance of Spatial
Multiplexed system for TSV channel model is simulated by
assuming a simple indoor Line of Sight (LOS) environment
model. Assuming perfect channel estimation, the Bit Error Rate
(BER) performance of the 2x2 system is investigated for Zero
Forcing (ZF), Minimum Mean Square Error (MMSE), and
Maximum Likelihood (ML) receivers.
Keywords-Millimeter Waves, Spatial Multiplexing, Triple Saleh
Valenzuela model, Zero Forcing, Minimum Mean Square Error,
Maximum Likelihood Equalizer.
I. INTRODUCTION
High data rates provided by the Millimeter Waves have
benefited the Wireless Communication Sector. To deal with
high network capacity of about Mbps or even more, the
network should have high bandwidth. The millimeter wave
band comes under an unlicensed spectrum band of 57-64 GHz
having a huge bandwidth of 7 GHz. The large available
bandwidth and O
2
absorption present at 60 GHz and other
intrinsic atmospheric attenuation make the millimeter waves
useful for high data rates and spatial reuse. The range of the
Millimeter Waves is limited by the blockage of waves by
walls and furniture, the link range for indoors being 10m and
outdoors 100m in the case of spatial reuse. This produces fast
and fading statistics of channel. At 60 GHz, free space path
loss is much higher. This loss as given by Friis Formula,
scales as
2
where = (C/f
c
), is the carrier wavelength and f
c
is the carrier frequency [1].
Modeling indoor propagation environment is complicated
by large variability in building layout and construction
materials. Another important element of indoor wireless
operation that should be taken into account is interference.
Indoor path loss can change dramatically with either time or
position, because of multipath present. The wideband of
waves used in indoor applications increase the sensitivity to
delay spread [2]. Site-specific and Site-general modeling are
the two general types of propagation modeling present. As the
path loss combines with other channel impairments like delay
spread, there arises a necessity to use directional antennas to
obtain reliable communications. To obtain better Signal to
Noise Ratio (SNR) in Millimeter wave band and in order to
effectively use frequency with Space Division method, the
effect of antenna directivity has to be considered[3],[4].
The Channel Model used in this paper for the Millimeter
Wave propagation is the Triple Saleh Valenzuela (TSV) which
is a modified form of Saleh Valenzuela (SV) model. This is a
merger of Two-Path model and SV model. This model was
contributed by NICT, Japan to 802.15.3c. The Complex
Impulse Response obtained from this model contains the Time
of Arrival information (TOA), Angle of Arrival information
(AOA), and the Complex amplitude of each ray in the cluster.
Thus by including the effect of directivity of antenna, the LOS
component is modeled along with NLOS components.
The capacity of the Multi Input Multi Output (MIMO)
channels increases linearly with the number of transmit and
receive antennas with no additional power expenditure. This
gain called the Spatial Multiplexing gain is obtained by
sending independent data from individual antennas [5]. The
Bit Error Rate performance of a 2x2 Spatial Multiplexed
system that uses TSV model in an indoor environment is
compared for different receivers. Maximum Likelihood
equalization technique was found to outperform ZF and
MMSE for both the cases.
The remainder of this paper is organized as follows. In
Section II, we describe the indoor environment considered and
the TSV channel model used. Section III describes the Spatial
Multiplexing in MIMO links and the different Equalization
techniques considered. Section IV deals with the performance
evaluation of the 2x2 Spatial Multiplexed BPSK systems with
ZF, MMSE, and ML equalization techniques.

II. TSV MODEL FOR INDOOR ENVIRONMENT
Indoor channel modeling is found to be difficult by the
large number of obstructions, variations in layout of the
978-1-61284-764-1/11/$26.00 2011 IEEE
building and movements of people. To satisfy all these, site-
general models are preferred. As the frequency of operation
for millimeter waves is very high, large bandwidths are
available for the link but this high frequency also leads to high
path loss.
The Triple Saleh Valenzuela model (TSV) is a site-general
model that merges S-V and two-path models. The impulse
response gives the information of the relative power of the
first ray that arrives to that of the last received signal
component. The amplitude factor of this impulse response in
TSV model is determined by the distance between the
millimeter device positions and the heights of the antennas.
The uncertainty and high vulnerability of the device position
and the fading caused is modeled here, in TSV by the random
variables generated by Poisson and Laplace distributions.
To consider the influence of antenna, the angle-of-arrival
information was included in the Impulse Response of
modified S-V model [7];
The Complex Impulse Response (CIR) of the TSV model is
given by [7] as
b(t) = [ o(t) + o
I,m
o (t -I
I
-
I,m
)

m=0

I=0
o( -
I
-
I,m
) (1)
Where, is the direct wave component that holds the
information about the heights of the transmitter and receiver
antenna, distance between the antenna, Reflection Co-efficient
and the wavelength of the center frequency.
o
I,m
is the complex amplitude of each ray.
t is the time, I
I
is the delay time of the l-th cluster,

I,m
is the delay time of the m-th ray in l-th cluster.

I
is the angle of arrival of the l-th cluster,
I,m
is the
angle of arrival of m-th ray in the l-th cluster.
Each ray belongs to the cluster. Probability of ray and
cluster generation is done by Poisson process and the
distribution of the angle is done by Laplacian distribution. The
positional parameters are treated as statistical and are
estimated by using Uniform distribution. Thus the equation (1)
is the modified form of Saleh Valenzuela model.
The direct wave component is expressed by the two-path
statistical model that helps to take into consideration the
uncertainties associated with the fading caused by slightest
movement of the device.
Layout Geometry as assumed in the Model is as follows
[7] as given by TSV model contributed by NICT:





TABLE 1: LAYOUT GEOMETRY ASSUMED IN THE TSV MODEL FOR
RESIDENTIAL ENVIRONMENT

Ceiling Height 2.47m
Window Height 2.11m
Transmitter Height 1.11m
Polarization Vertical
Transmitter Antenna Always Fixed
Receiver Antenna Rotated from 0
0
to 360
0
in steps of
5
0
.
Assumed Distance 5m
Environment Used Residential

III. SPATIAL MULTIPLEXING FOR LOS
By providing multi transmit and multi receive antennas,
spatial dimension increases and this leads to the increase to the
increase in degree-of-freedom gain. This can be further
exploited by spatially multiplexing the streams of data which
leads to the increase in capacity or data rate of the MIMO
channel.
Considering a MIMO structure containing {1,2,.M
T
}
transmitter antennas and {1,2,M
R
} receiver antennas , the
discrete time channel is modeled using the equation,[8]
= EX +N (2)
Where, Y is the (M
R
x 1) channel output, X is the (M
T
x1)
transmitted data. H is the (M
T
x M
T
) impulse response of the
channel which contains the TOA and AOA information and N
is the (M
R
x 1) Additive White Gaussian Noise (AWGN).
Spatial Multiplexing for MIMO links can be achieved for
moderate antenna spacing even with a less scattering
environment. The objective of the Spatial Multiplexed systems
as opposed to Space-Time diversity coding is to maximize the
transmission rate and thus the spectral efficiency [9].
Accordingly, the M
T
independent data symbols are transmitted
per symbol period. At high SNR, the channel capacity
increases with SNR as min {M
T
, M
R
} log SNR (bps/Hz), in
contrast to log SNR for single channels. Thus multiple antenna
channels are min {M
T
, M
R
} parallel spatial channels, hence it
is the total number of degrees of Freedom to communicate
[10].
Any signal processing technique that is used to mitigate
the Inter Symbol Interference (ISI) that is caused by the delay
spread is called Equalization. When the signal from the
channel passes through the equalizer, it increases the noise
power also. Hence, proper techniques should be used to reduce
the noise enhancement.
Equalizers fall under two categories; Linear and Non-
Linear. Non-linear equalizers have less noise enhancement
than linear equalizers but have high complexity
The remainder of this section deals with the Zero Forcing,
Minimum Mean Square Error and Maximum Likelihood
equalization techniques at the receiver.
A. Maximum Likelihood Receiver (ML)
The transmitter sends one of M signals s
i
(t), for
i=1,2,,M. The M signals forms a constellation in the
signaling space. The received signal y(t) = s
i
(t) + n(t) is
decomposed to its components in the signal space. The ML
receiver performs optimum vector decoding and is optimal in
the sense of minimizing the error probability [11]. The ML
receiver compares the received signal with all the transmitted
signals which would be modified by the channel matrix H and
finds the transmit symbol vector x by the equation given as.,

y = org
x
k e {j
1
,j
2,
j
N
]
min [ r - E

k
[
2
(3)
where the minimization is performed over all possible
transmit estimated vector symbols y and H is the (M
T
x M
T
)
impulse response of the channel which contains the TOA and
AOA information. Although ML detection offers optimal error
performance, it suffers from complexity issues. The maximum
Likelihood receiver picks the signal that is closed to the
received signal in the signal space. It has exponential
complexity in the sense that the receiver has to consider |C|
M

possible symbols for an M transmitter antenna system with C
as the modulation constellation.
B. Zero Forcing Receiver
Considering the MIMO channel model given in (2), where
the N data sub streams are mixed by the channel matrix. The
ZF equalizer can be applied to decouple the N sub streams.
[12], [16].
W
zf
= (H*H)
-1
H* (4)
where H is the (M
T
x M
T
) impulse response of the channel
which contains the TOA and AOA information and H
*
is the
conjugate of H.
Multiplying the received signal vector y on the Left Hand
Side by W
zf
, N decoupled sub streams is obtained with output
SNRs given as.,
p
z,] n
=
sn
(H
-
H)
-1
nn
, 1 n N. (5)
The ZF receiver converts the joint decoding problem into
M single stream decoding problems thereby significantly
reducing receiver complexity. This leads to the trade-off
between complexity reduction and performance degradation.
[13].
C. Minimum Mean Square Error(MMSE)
Considering the MIMO channel model given in (2), where
the N data substreams are mixed by the channel matrix. The
MMSE equalizers can be applied to decouple the N
substreams.[14].,
WMMSE = [E
-
E +
1
sn
I
-1
H*. (6)
Multiplying the received signal vector y on the Left Hand
Side by W
MMSE
, N decoupled substreams is obtained with
output SNRs given as.,
p
mmsc,n
=
sn
[H
-
H+
1
snr
I
-1
nn
-1, 1 n N. (7)
The MMSE receiver suppresses both the interference and
noise components, whereas the ZF receiver removes only the
interference components. Some of the important
characteristics of MMSE detector are simple linear receiver,
superior performance to ZF and at Low SNR, MMSE becomes
matched filter. The linear ZF and MMSE equalizers are classic
functional blocks and are ubiquitous in digital
communications [15]. They are also the building blocks of
more advanced communication schemes such as the decision
feedback equalizer (DFE), or equivalently, the V-BLAST
(vertical Bell Labs layered Space-Time) architecture and
various other MIMO transceiver designs. It is commonly
understood that ZF is a limiting form of MMSE as
Eb/N0 - [8]. Eb/N0 is the energy per bit to noise power
spectral density ratio. It is normalized Signal to Noise Ratio
(SNR) measure. It is equal to SNR divided by link spectral
efficiency in (b/s)/Hz. It is used in power-limited rather than
bandwidth-limited communications.
IV. RESULTS AND DISCUSSION
TABLE 2: SIMULATION RESULTS OF THE TSV MODEL IN
RESIDENTIAL ENVIRONMENT

LOS Component Pathloss -81.9842[dB]
Average RMS delay 1.290 [ns]
Maximum RMS delay 2.657 [ns]
Minimum RMS delay 0.693 [ns]
Average Rician factor 27.023 [dB]
Maximum Rician factor 35.799 [dB]
Minimum Rician factor 14.737 [dB]
Environment Residential
Distance assumed between
Transmitter and Receiver
5m
Number of clusters 4











Figure 1. Power Delay Profile showing LOS and NLOS component
0 20 40 60 80 100 120 140
-140
-130
-120
-110
-100
-90
-80
A
v
e
r
a
g
e

p
o
w
e
r

[
d
B
]
Power delay profile
The Power Delay Profile (PDP) in Fig.1shows the presence of
direct component having the Average Power of nearly -80 dB.
As the number of reflections per ray increases, the
corresponding amplitude of the ray decreases, because of both
reflection losses and higher free space losses. Hence, the ray
amplitude depends on room dimensions and the magnitude of
reflection co-efficients.















Figure 2. Complex Impulse Response as a function of Relative Power
and Time of arrival.













Figure. 3. Complex Impulse Response as a function of Time of Arrival
and Angle of Arrival information.
Power Delay Profile in Figures 2and 3 contains the AOA
information which includes the effect of antenna directivity
that helps to obtain SNR characteristic of Millimeter Wave
band and the effective use of frequency in Space-Division
method. In this, we can observe the ray clustering which is
evidenced by the peaks in the PDP.
TABLE 3: SIMULATION SUMMARY FOR BIT ERROR RATE
PERFORMANCE FOR 2X2 SPATIAL MULTIPLEXED SYSTEM.

Modulation Used BPSK
Eb/N0 1:15dB
Channel Model Triple Saleh Valenzuela
Environment Used Residential
Centre Frequency 60GHz
Bit Error Rate obtained 10
-4

















Figure.4. Bit Error Rate performance for a 2x2 Spatial Multiplexed system
From the Fig.4, it can be inferred that the ML receiver is
optimal in reducing the error probability and estimates better as
it compares the received signal with all the possible transmitted
vectors while the Zero Forcing receiver has the advantage of
complexity reduction, but as the multipath components are
large, the noise components are not suitably taken care of by
ZF, hence ZF shows a reduced performance. MMSE technique
is able to reduce both the interference and noise components as
compared to ZF eventhough it has a slight edge in
performance with 2x2 antennas, as the number of antennas
increase, they perform almost equally.
0 5 10 15 20 25
-140
-130
-120
-110
-100
-90
-80
Time of arrival[ns]
R
e
l
a
t
i
v
e

p
o
w
e
r

[
d
B
]
Power delay profile


clus 1
clus 2
clus 3
clus 4
100
150
200
250
-100
0
100
200
-20
0
20
40
60
80

Power delay profile
Angle of arrival [deg]
1
0
*
l
o
g
1
0
(
a
b
s
(
a
m
p
l
i
t
u
d
e
) 2
)
+
1
0
0

[
d
B
]
clus 1
clus 2
clus 3
clus 4
2 4 6 8 10 12 14
10
-6
10
-5
10
-4
10
-3
10
-2
10
-1
10
0
Eb/No (dB)
B
E
R
2x2 BPSK System


ZF-SIC
MMSE-SIC
ML-SIC
V. CONCLUSION
In this paper, the BER performance of 2x2 Spatial
Multiplexed system is studied in indoor environment
parameters using a Triple Saleh Valenzuela model. The
parameters considered for the simulation of the Triple Saleh
Valenzuela model is as contributed by NICT Japan to the
TG3c group. The simulation was performed for an indoor
Residential environment having a distance of 5m considering
the transmitter antenna beamwidth as 60
0
and receiver antenna
beamwidth as 30
0
and the Average Delay Spread is found to
be 1.290ns.Considering the Spatial Multiplexing, the data rate
is obtained as 2.2Gbps and the BER performance is found to
be better when the number of transmit and number of receive
antennas were increased.
REFERENCES
[1] Millimeter Wave Technology in Wireless PAN, LAN,and MAN,
Edited by Shao-Qiu Xiao, Ming-Tuo Zhou,Yan Zhang, CRC Press,
Auerbach Publications, ISBN-13:978-0-8493-8227-7,2008
[2] A. Poon, R. Brodersen, and D. Tse, Degrees of freedom in multiple
antenna channels: A signal space approach, Information Theory,
IEEE Transactions on, vol. 51, no. 2, pp. 523 536, Feb. 2005.
[3] Andre Pollok, William G. Cowley, and Ian D. Holland, Multiple-
Input Multiple-Output Options for 60 GHz Line-of-Sight Channels,
AusCTW 2008, pp.101-106, 2008.
[4] Eric Torkildson, Colin Sheldon, Upamanyu Madhow, and Mark
Rodwell, Millimeter-Wave Spatial Multiplexing in an Indoor
Environment.
[5] Sylvain Ranvier, Suiyan Geng and Pertti Vainikainen, Mm-Wave
MIMO Systems for High Data-Rate Mobile Communications,
Wireless VITAE09, pp. 142-146, 2009.
[6] S. K. Yong and C. C. Chong, An overview of multigigabit wireless
through Millimeter wave technology: potentials and technical
challenges, EURASIP Journal on Wireless Communications and
Networking, no. 1, pp. 1-10, vol. 2007.
[7] Hirokazu Sawada, Yozo Shoji and Chang-Soon Choi, Proposal of
novel statistic channel model for Millimeter Wave WPAN TSV:
Sawada-Saleh (Triple S) Valenzuela model,Proceedings of Asia-
Pacific Microwave Conference 2006.
[8] Arogyaswami J. Paulraj, Dhananjay, A. Gore, Rohit U. Nabar, and
Helmut Blcskei, An Overview of MIMO CommunicationsA Key
to Gigabit Wireless, Proceedings Of The IEEE, Vol. 92, No. 2,
pp.198- 218, February 2004.
[9] C. Liu, E. Skafidas, and R. J. Evans, Capacity and Data Rate for
Millimeter Wavelength Systems in a Short Range Package Radio
Transceiver ,IEEE Transactions On Wireless Communications, Vol.
9, No. 3, pp.902- 906, March 2010.
[10] Colin Sheldon, Eric Torkildson, Munkyo Seo, C. Patrick Yue, Mark
Rodwell, and Upamanyu Madhow, Spatial Multiplexing Over a
Line- of-Sight Millimeter-Wave MIMO Link: A Two-Channel
Hardware Demonstration at 1.2Gbps Over 41m Range, Proceedings
of the 1st European Wireless Technology Conference, pp. 198
221, October 2008.
[11] G. J. Foschini and M. Gans, On the limits of wireless
communications in a fading environment when using multiple
antennas, Wireless Personal Communications, vol. 6, no.3, p. 311,
Mar. 1998.
[12] P. Wolniansky, G. Foschini, G. Golden, and R. Valenzuela, "V-
BLAST: an architecture for realizing very high data rates over the
rich-scattering wireless channel," in Proc. ISSSE98, 1998, pp. 295-
300.
[13] P. F. M. Smulders, Deterministic modelling of indoor radio
propagation at 40-60 Ghz, Wireless Personal Communications, vol.
4, no. 2, pp.127135, June 1994.
[14] C. Sheldon, M. Seo, E. Torkildson, M. Rodwell, and U. Madhow,
Four channel spatial multiplexing over a millimeter-wave line-of-
sight link, IEEE - MTTS International Microwave Symposium,
June 2009.
[15] F. Bohagen, P. Orten, and G. Oien, Construction and capacity
analysis high-rank line-of-sight MIMO channels, IEEE Wireless
Communications and Networking Conference,vol.1, pp. 432
437,Mar. 2005.
[16] I. Sarris, A. R. Nix, and A. Doufexi, High performance wimax
architecture using MIMO technology in line-of-sight, The Second
European Conference on Antennas and Propagation, pp. 15, Nov.
2007.