2003-05-08
Project
Title
C802.20-03/50
IEEE 802.20 Working Group on Mobile Broadband Wireless Access
<http://grouper.ieee.org/groups/802/20>
Overview of METRA Model for MBWA MIMO Channel
Date
Submitted
2003-05-08
Source(s)
Insoo Sohn
ETRI
Voice: +82-42-860-1027
Fax: +82-42-860-6732
Email: [email protected]
Glenn D. Golden
Flarion Technologies, Inc.
Voice: 908-997-2000
Fax: 908-947-7090
Email: [email protected]
Heesoo Lee
ETRI
Voice: +82-42-860-5375
Fax: +82-42-860-6732
Email: [email protected]
Jae Young Ahn
ETRI
Voice: +82-42-860-6421
Fax: +82-42-860-6732
Email: [email protected]
Re:
Abstract
This contribution presents the METRA MIMO channel model as the MIMO channel model
for MBWA system. MIMO channel parameters are also given as an example.
Purpose
For informative use only.
Release
This document has been prepared to assist the IEEE 802.20 Working Group. It is offered as a basis for
discussion and is not binding on the contributing individual(s) or organization(s). The material in this
document is subject to change in form and content after further study. The contributor(s) reserve(s) the right
to add, amend or withdraw material contained herein.
The contributor grants a free, irrevocable license to the IEEE to incorporate material contained in this
contribution, and any modifications thereof, in the creation of an IEEE Standards publication; to copyright in
the IEEE’s name any IEEE Standards publication even though it may include portions of this contribution;
and at the IEEE’s sole discretion to permit others to reproduce in whole or in part the resulting IEEE
Standards publication. The contributor also acknowledges and accepts that this contribution may be made
public by IEEE 802.20.
Patent
Policy
The contributor is familiar with IEEE patent policy, as outlined in Section 6.3 of the IEEE-SA Standards
Board Operations Manual <http://standards.ieee.org/guides/opman/sect6.html#6.3> and in Understanding
Patent Issues During IEEE Standards Development <http://standards.ieee.org/board/pat/guide.html>.
Notice
1
2003-05-08
C802.20-03/50
Overview of METRA Model for MBWA MIMO Channel
1 Introduction
This document provides a brief introduction to the Multiple Element Transmit Receive Antenna
(METRA) technique for modeling Multiple Input/Multiple Output (MIMO) channels. This technique,
and the specific models derived from it, will be of interest to the channel modeling community of the
Mobile Broadband Wireless Access system (MBWA).
2 MIMO Channels
2.1 Importance of the MIMO Channel
Communication systems employing multiple transmit and multiple receive antenna elements, known as
MIMO systems, are gaining considerable interest for next generation mobile wireless communication
systems. Theoretical work by Foschini and Gans [1] has shown that in sufficiently rich scattering
environments, MIMO holds the potential for enormous spectral efficiency – hence capacity -improvements relative to SISO systems. Conceptually, such capacity gains result from exploiting the
multiple parallel sub-channels created by the scattering decorrelation. Recently, spectral efficiencies of
20-40 b/s/Hz (in 30 kHz bandwidth) were demonstrated at Lucent using Vertical-Bell Laboratories
Layered Space Time (V-BLAST) system in an indoor office environment [4]. Fig. 1 shows a MIMO
system setup with N antennas at the mobile station and M antennas at the base station.
2.2 MIMO Modeling Approaches
There are three main approaches to MIMO channel modeling: The correlation or METRA model, the raytracing model, and the scattering model. The properties of these models are briefly as follows:
 METRA Model: This model characterizes spatial correlation by combining independent complex
Gaussian channel matrices at the transmitter and receiver [2,3]. For multipath fading , the ITU model is
used to generate the power delay profile and Doppler spectrum. Since this model is based on ITU’s
generalized tap delay line channel model, the model is simple to use and backward compatible with
existing ITU channel profiles.
 Ray-Tracing Model: In this approach, exact locations of the primary scatterers are assumed known.
The resulting channel characteristics are then predicted by summing the contributions from a large
number of the paths thru the simulated environment from each transmit antenna to each receive antenna.
2
2003-05-08
C802.20-03/50
This technique provides fairly accurate channel prediction; however, it is too complex for major outdoor
environment model.
 Scattering Model: This model assumes a particular statistical distribution of scatterers. Using this
distribution, channel models are generated through simulated interaction of scatterers and planar
wavefronts. The disadvantage of this model is that it requires a large number of parameters.
Base Station (BS)
Mobile Station (MS)
s1 (t )
y1( t )
s2 (t )
y 2 (t )
s3 (t )
Scattering
Medium
s N (t )
y3 ( t )
y M (t )
Figure 1: MIMO System in a Scattering Environment
2.3 METRA Channel Model
2.3.1 METRA Project
The METRA project is one of the various projects under the Information Society Technologies (IST)
Program managed by the Information Society Directorate-General of the European Commission (EC).
The METRA consortium consists of Universitat Politecnica de Catalunya, Aalborg University, Nokia
Networks, Nokia Mobile Phones, and Vodafone Ltd. The METRA project is based on previous EC
funded projects on smart antennas for mobile communications such as Technology in Smart Antennas for
Universal Advanced Mobile Infrastructure (TSUNAMI) Project and Smart Universal Beamforming
(SUNBEAM) project.
2.3.2 Channel Characterization
The procedure for generating MIMO channel characteristics from the METRA model is shown in Fig. 2.
3
2003-05-08
C802.20-03/50
The procedure is divided into two major phases. In the first phase, a correlation matrix is generated for
each mobile station (MS) and base station (BS) based on the number of antennas, antenna spacing,
number of clusters, power azimuth spectrum (PAS), azimuth spread (AS), and angle of arrival (AoA).
These two correlation matrices are combined to create a spatial correlation matrix using the Kronecker
product. In the second phase, a correlated signal matrix is created using fading signals derived from
various Doppler spectra and power delay profiles, and a symmetrical mapping matrix based on the spatial
correlation matrix. Some of the parameters that can be used in the METRA channel model are shown in
table 1 (source: 3GPP TR 25.876 V1.0.1). Correlation matrices generated using the METRA model for a
4x4 system (4 transmit and 4 receive antennas) in the ITU Pedestrian A and Vehicular A environments
are shown in Fig. 3. Fig. 4 shows channel gains generated using the METRA model for a 2x2 system in
the Pedestrian A environment. Finally, the Classical Doppler spectrum for a 2x2 system is shown in
Fig. 5.
2.3.3 Validation of METRA Modeling Approach
The METRA modeling approach has been validated by comparing the generated models with measured
data obtained in both the picocell and microcell environments [2]. In a 4x4 microcell environment,
vertically polarized dipole transmit antenna on a rotating bar with a circumferential distance of 20  and
uniform linear receive antenna array were deployed in small office environment. Another 4x4 microcell
environment was also evaluated, using interleaved transmit antenna array with four vertically polarized
sleeve dipoles elements moving along a linear slide with 11.8  and uniform linear receive antenna array,
also in a small office environment.
The model was validated using eigen-analysis methods.
The
eigenvalues estimated from the measured data closely matched the eigenvalues generated from the
METRA model.
4
2003-05-08
C802.20-03/50
num. of antenna
antenna spacing
cluster
PAS
AOA, AS
BS and MS correlation
matrix
Phase 1
Spatial correaltion
matrix generation
R, Q, sigma
Spatial correaltion
matrix
R=kron(RBS,RMS) : down
R=kron(RMS,RBS) :up
Symmetrical mapping
matrix
C=chol(R)
Phase 2
Correlated fading
signal generation
Uncorrelated
fading signal
a
x
correlated signal matrix
A=sqrt(P)Ca
Figure 2: METRA MIMO Channel Modeling Procedures
5
2003-05-08
C802.20-03/50
Table 1: MIMO Channel Parameters
Case A
Rayleigh
Uncorrelaed
Case B
Macrocell
Ped A
Case C
Macrocell
Veh A
Case D
Microcell/Bad-urban
PedB
Number of paths
1
4
6
6
PDP
N/A
ITU Pedestrian A
ITU Vehicular A
ITU Pedestrian B
Laplacian
Doppler spectrum
Classical
Classical
Laplacian
Speed(km/h)
3/40/120
3/40/120
3/40/120
3/40/120
Topology
N/A
0.5 spacing
0.5 spacing
0.5 spacing
PAS
N/A
Path #1, Rician, K=6dB
Uniform over 360
Laplacian, AS=35
(Uniform over 360)
Laplacian, AS=35
(Uniform over 360)
DoM(deg)
N/A
0
22.5
-22.5
67.5 (all path)
22.5 (odd paths)
-67.5 (even paths)
UE
AoA(deg)
N/A
22.5
Topology
N/A
PAS
N/A
Laplacian, AS=5
Laplacian, AS=10
Laplacian, AS=15
N/A
20,501)
20,501)
2,-20,10,-8,-33,312)
Uniform linear array:
1) 0.5 spacing
2) 4.0 spacing
Node B
AoA
Case B (Pedestrian A)
1
0.4640+0.8499i -0.4802+0.7421i -0.7688-0.0625i 

 0.4640-0.8499i
1
0.4640+0.8499i -0.4802+0.7421i 

 -0.4802-0.7521i 0.4640-0.8499i
1
0.4640+0.8499i 


1
 -0.7688+0.0625i -0.4802-0.7421i 0.4640-0.8499i

 1 -0.3043 0.2203 -0.1812 
-0.3043 1
-0.3043 0.2203

 0.2203 -0.3043
1
-0.3043 


1
-0.1812 0.2203 -0.3042

Node B
Laplacian, AS=5,
0.5 , AOA=20
UE
Uniform, K=6dB
0.5 , AOA=22.5
Case C (Vehicular A)
1
0.4290+0.7766i -0.3642+0.5475i -0.4527-0.0502i 

 0.4290-0.7766i
1
0.4290+0.7766i -0.3642+0.5475i 

 -0.3642-0.5475i 0.4290-0.7766i
1
0.4290+0.7766i 


1
 -0.4527+0.0521i -0.3642-0.5475i 0.4290-0.7766i

1
-0.6906+0.3419i 0.4903-0.3626i -0.3733+0.3450i 

 -0.6906-0.3419i
1
-0.6906+0.3419i 0.4903-0.3626i 

 0.4903+0.3626i -0.6906-0.3419i
1
-0.6906+0.3419i 


1
 -0.3733-0.3450i 0.4903+0.3626i -0.6906-0.3419i

Node B
Laplacian, AS=10,
0.5 , AOA=20
UE
Laplacian, AS=35, DOM=22.5
0.5 , AOA=67.5
Figure 3: METRA Model Correlation Matrices
6
2003-05-08
C802.20-03/50
Tx#1 - Rx#1
1
Tx#1 - Rx#2
1
10
10
0
10
0
10
-1
10
-1
10
-2
10
-2
10
-3
10
-4
10
-3
0
0.02
0.04
0.06
0.08
10
0.1
0
0.02
0.04
Time [s]
Tx#2 - Rx#1
1
0.08
0.1
0.08
0.1
Tx#2 - Rx#2
1
10
10
0
0
10
10
-1
-1
10
10
-2
-2
10
10
-3
-3
10
10
-4
10
0.06
Time [s]
-4
0
0.02
0.04
0.06
0.08
10
0.1
0
0.02
0.04
Time [s]
0.06
Time [s]
Figure 4: METRA Model Channel Gain with 2Tx and 2Rx Antennas
Tap h111
Tap h211
1
0.1
0.8
0.08
0.6
0.06
0.4
0.04
0.2
0.02
0
-2
-1
0
Tap h112
1
2
0
-2
1
0.1
0.8
0.08
0.6
0.06
0.4
0
Tap h121
1
2
0
-2
0.05
0.8
0.04
0.6
0.03
0.4
2
0.015
0.015
0.01
0.01
0.005
0.005
0
-2
-1
0
Tap h312
1
2
0
-2
0.04
0.02
0.03
0.015
0.02
0.01
0.01
0.005
-1
0
Tap h412
1
2
-1
0
Tap h221
1
2
0
-2
-1
0
Tap h321
1
2
0
-2
0.02
0.02
-1
0
Tap h421
1
2
0.015
0.015
0.01
0.01
0.005
0.005
-1
0
Tap h422
1
2
-1
0
1
2
0.02
0.2
0.01
-1
0
Tap h122
1
2
0
-2
1
0.1
0.8
0.08
0.6
0.06
0.4
0.04
0.2
0.02
0
-2
1
0.02
-1
1
0
-2
0
Tap h212
Tap h411
0.02
0.04
0.2
0
-2
-1
Tap h311
0.02
-1
0
1
2
0
-2
-1
-1
0
Tap h222
0
1
2
1
2
0
-2
-1
0
Tap h322
1
2
0
-2
0.04
0.02
0.03
0.015
0.02
0.01
0.01
0.005
0
-2
-1
0
1
2
0
-2
Figure 5: METRA Model Doppler Spectrum with 2TX and 2RX Antennas
7
2003-05-08
C802.20-03/50
2.4 Performance
To demonstrate the performance improvements possible with MIMO systems the METRA approach was
used to generate MIMO channel models for a MIMO-OFDM system. This system has 2 transmit and 2
receive antennas and uses space-frequency block coding (SFBC) and spatial division multiplexing (SDM)
for capacity enhancement. A low-density parity check code (LDPC) was used for channel coding. The
METRA channel with the Case B parameters from Table 1 and mobile speed of 60 km/h was used. The
packet error rate (PER) performance for 2x1 and 2x2 SFBC schemes and a single antenna (SISO) scheme
as a reference are shown in Fig. 6. It can be seen that there is around 3 dB gain for 2 Tx-1 Rx SFBC and
around 5dB gain for 2Tx-2Rx SFBC compared to the single antenna scheme for PER = 10-3. Fig. 7 shows
the PER performance for SDM which achieves double capacity enhancement.
0
10
(1,1) Single Antenna
(2,1) SFBC
(2,2) SFBC
-1
PER
10
-2
10
-3
10
-4
10
4
6
8
10
12
14
16
18
20
22
Average Es/No (dB)
Figure 6: Simulation Results for MIMO-OFDM with SFBC
0
10
(2,2) SFBC
(2,2) SDM
-1
PER
10
-2
10
-3
10
-4
10
4
6
8
10
12
14
16
18
20
Average Es/No (dB)
Figure 7: Simulation Results for MIMO-OFDM with SDM
8
22
2003-05-08
C802.20-03/50
3 Summary
MIMO systems have the potential for significant capacity improvements for MBWA, relative to
traditional SISO and diversity techniques. In order to exploit these potential gains, high quality MIMO
channel models are required. The METRA approach provides a practical and accurate technique for
modeling MIMO channels.
References
[1] G.J. Foschini, M.J. Gans, “On limits of Wireless Communications in a Fading Environment when
using Multiple Antennas,” Wireless Personal Communications, No. 6, 1998, pp.311-335
[2] J. P. Kermoal, L. Schumacher, K. I. Pedersen, P. E. Mogensen, and F. Frederiksen, “A Stochastic
MIMO Radio Channel Model with Experimental Validation,” IEEE JSAC, vol. 20, pp. 1211-1225, Aug.
2002.
[3] L. Schumacher, J. P. Kermoal, F. Frederiksen, K. I. Pedersen, A. Algans, and P. E. Mogensen, “MIMO
Channel Characterisation,” IST Project IST-1999-11729 METRA Deliverable D2, Feb. 2001
[4] G. D. Golden, G. J. Foschini, R. A. Valenzuela, and P. W. Wolniansky, “Detection Algorithm and Initial
Laboratory Results using V-BLAST space-time Communication Architecture”, Electronics Letters, Jan. 7,
1999, Vol. 35, No. 1.
9