Smart Antennas for Mobile Communication Systems - ZID ...

Project Report
18.12.1998
"Smart Antennas for Mobile Communication Systems"
Project Leader: Prof. Ernst Bonek
Report Editors: Dipl.-Ing. Juha Laurila
Dipl.-Ing. Klaus Hugl
Technische Universität Wien
Institut für Nachrichtentechnik und Hochfrequenztechnik

P-12147 MAT Project Report 1997-98
1 INTRODUCTION 3
2 PROJECT DOCUMENTATION 4
2.1 Channel Modeling 4
2.1.1 Introduction 4
2.1.2 Parameter fitting of the GSCM 4
2.1.3 Time variation of the spatial channel model 5
2.1.4 Downlink channel model 5
2.2 Downlink Algorithms 6
2.2.1 Introduction 6
2.2.2 Estimation of the spatial covariance matrix 6
2.2.3 Null Broadening 7
2.2.4 Beamforming Algorithms 8
2.2.5 Downlink Transmit Power 8
2.2.6 Simulations and Results 9
2.3 Receiver Imperfections and Calibration 9
2.3.1 Introduction 9
2.3.2 Model of Imperfections 10
2.3.3 Simulations and Results 11
2.3.4 Calibration Considerations 12
2.4 Robust Algorithms (Blind and Semi-Blind Estimation) 13
2.4.1 Introduction 13
2.4.2 Subspace Estimation (Space-Time Equalisation) 14
2.4.3 DILSF Algorithm 14
2.4.4 Simulations and Results 15
2.4.5 Testbed Software Implementation 16
2.5 References 17
3 PROJECT RELATED PUBLICATIONS 20
"Smart Antennas for Mobile Communications Systems" 2

P-12147 MAT Project Report 1997-98
1 INTRODUCTION
This document summarises the research activities carried out in the framework of the project "Smart
Antennas for Mobile Communication System". The project is based on the contract P-12147 MAT
between Austrian Science Fund (Fonds zur Förderung der wissenschaftlichen Forschung, FWF) and
Institut für Nachrichtentechnik und Hochfrequenztechnik, TU Wien (INTHF). The project started in
May 1997 and this deliverable considers the achievements during the first two project years.
The aim of the project is to contribute to the utilisation of adaptive antennas with the existing and
forthcoming mobile communication systems. The focus of the research is to develop new signal
processing schemes for adaptive antennas in mobile radio application. However, we have also
carefully considered application specific issues like influences and restrictions caused by physical
propagation phenomena and imperfect receiver hardware.
This deliverable omits technical details, but gives an overview and acts as a guide to find more
information in the other project related publications with more technical nature. The section is divided
further into subsections, each describing one of the following main activities:
• Channel Modeling
• Downlink Algorithms
• Receiver Imperfections and Calibration
• Robust Algorithms (Blind and Semi-Blind Estimation)
"Smart Antennas for Mobile Communications Systems" 3

P-12147 MAT Project Report 1997-98
2 PROJECT DOCUMENTATION
2.1 Channel Modeling
2.1.1 Introduction
Realistic spatial channel modeling is vital for the simulations of adaptive antenna systems. Only a
channel model which reflects the radiowave propagation in typical environments, in the time as well
as in the angular domain, can be used for simulations. Violation of this basic legitimacy may lead to
wrong conclusions about the analysed algorithms and the system itself.
The theoretical basis of the Geometry-based Stochastic Channel Model (GSCM) [FMB_98] was
already existing before the current project. The model consists of scatterers 1 in the vicinity of the
mobile station and the received signal is calculated by simple ray-tracing. The location of the
scatterers is selected according to the appropriate PDF (probability density function). In addition, it is
possible to include far scatterers which correspond to reflections from far obstacles, like mountains in
rural environments or high-rise buildings in suburban and urban environments.
In the framework of the current project we extended and parametrised the model to enable realistic
simulations with different purposes and environments. For example, we have included the MS
movement to be able to consider the time variation of the channel. Further, we had to fit the
parameters of the model in a way that the characteristics like power delay profile (PDP) and azimuthal
power spectrum (APS) agree with recent channel sounder measurements [Ped_97]. In addition to the
modifications of the uplink channel model we created a corresponding downlink implementation with
the same requirements.
2.1.2 Parameter fitting of the GSCM
The fundamental requirement for the spatial channel models is that the PDP and APS correspond to
the distributions observed in the channel measurements. As discussed above, the distribution defining
the location of the scatterers forms the basis of the GSCM and determines also the responses at the
receiving antennas in a straightforward way. In our model we have considered three different PDFs for
the distance between the MS and the scatterers; Gaussian, Rayleigh and uniform distributions
[LSB_98]. Figure 1 shows the simulated azimuthal power spectra with these PDFs when only local
scatterers were present. Comparison to the recent measurements results [Ped_97] shows that Gaussian
distribution fits well to the measured Laplacian APS and it approximates properly also the exponential
PDP of the well known stochastic COST 207 models. Corresponding considerations have been carried
out also for the scenario with far scatterers. Our results show that parameter selection modeling urban
environment with base stations above the rooftop level gives perfect fit with the COST 207 TU
(typical urban) model (Fig. 3.)
The investigations shown above are also possible by employing analytical bijective mathematical
transformation between scatterer locations and resulting delays and DOAs (direction of arrival) at the
receiver. Accordingly the measured joint ADPS (azimuthal delay power spectrum) can be transformed
to the scatterer constellation under the single scattering assumption [MLK_98], [Egg_98]. This
enables direct parametrisation of the GSCM by utilising measured data.
1
Following the established notation in the literature we use word "scatterer" even if strictly speaking specular
reflections are assumed.
"Smart Antennas for Mobile Communications Systems" 4

P-12147 MAT Project Report 1997-98
2.1.3 Time variation of the spatial channel model
In normal propagation environments buildings, hills and mountains act as the dominant reflectors.
Therefore the scattering points do not change their position when the MS moves through the scattering
scenario. Further, in link level simulations we are typically interested in the short term variations of
the mobile radio channel. For downlink beamforming we have to carry out the parameter estimation
by means of averaging, because the instantaneous downlink channel is not known due to the
uncorrelated fading in different transmission directions. However, averaging over small-scale fading is
adequate which corresponds to the mobile movement over 10-20λ. Notation λ denotes the wavelenght
of the signal, being about 16 cm for GSM1800 frequencies.
However, for some special simulation purposes it might be also relevant to consider longer term
channel variations, which correspond to the movement of the MS over the hundreds of wavelenghts.
For these situations we have proposed the special implementation of the GSCM, in which we locate
scatterers uniformly over the whole area under consideration beforehand and we weight the scattering
cross sections according to the MS movement during the simulation. We call this implementation
NSCS, nonuniform scatterer cross section [MKLH_98]. Additionally, we have also proposed to
include other long-term effects, like shadowing, if needed.
Normalised Power [dB]
0
−2
−4
−6
−8
−10
−12
−14
−16
−18
Power Azimuth Spectra
−20
−8 −6 −4 −2 0 2 4 6 8
Azimuth around the nominal DOA
Fig. 1. APS for the GSCM Fig. 2. PDP for the GSCM
2.1.4 Downlink channel model
Rayl.−analytical
Rayl.−simulation
Uniform
Gaussian
−30
0 1 2 3 4 5 6 7
Delay [us]
Our downlink channel model has identical principle than the corresponding uplink implementation
described above. Naturally, also the scattering points have same locations. The only required
modifications for the downlink model were the time displacement between uplink and downlink
transmission (e.g. three timeslots in GSM system) and different carrier and Doppler frequencies. The
frequency shift between uplink and downlink, however, leads to fully uncorrelated channel responses.
This modified uplink channel model and the corresponding downlink part were used in all the
simulations described in this document.
"Smart Antennas for Mobile Communications Systems" 5
Normalised power [dB]
0
−5
−10
−15
−20
−25
Power Delay Profile − local & far scatterers
Our Model
COST 207 TU Profile

P-12147 MAT Project Report 1997-98
2.2 Downlink Algorithms
2.2.1 Introduction
Uplink array processing is different to that of the downlink in an adaptive antenna system. In uplink
reception we can react on the actual channel situation and several algorithms are proposed in literature
(see overview in [Fuhl_97]) which promise large capacity gains.
But in the downlink the channel is unknown. We have to extract the important parameters of the
uplink channel to get an estimate of the downlink propagation situation if we want to prevent mobile
station feedback [GP_96]. But the fading in up- and downlink is totally uncorrelated due to the
different carrier frequencies in an FDD (Frequency Division Duplex) system. In downlink
transmission the fading is unknown and therefore the unique possibility is to utilise the mean channel
characteristics. Only the directions and the mean powers of the signal paths are the same in up- and
downlink. As a consequence the gain of an adaptive antenna system is smaller by about 3-4dB in the
downlink.
The goal of a mobile communication system applying adaptive antennas is to get the same capacity
increase in the downlink as in the uplink. Therefore it is essential to develop signal processing
methods which can cope with the problems of the downlink. Moreover the importance of the downlink
part has increased dramatically in the last few years. Data transmission applications of cellular mobile
communication systems, e.g. the Internet, require a higher downlink capacity compared to the uplink.
In the downlink our goal is to select the weight vectors in a way that most of the power is transmitted
to the desired user and the produced interference for the other users is kept as low as possible. We
divide the weight calculation in four consecutive steps which are explained in the sequel. Our
downlink beamforming approach is illustrated in Fig. 1.2.1.
Covariance
Matrix
Estimation
Fig. 3. Consecutive steps of the downlink beamforming process.
2.2.2 Estimation of the spatial covariance matrix
We used the spatial covariance matrix of the downlink channel for weight calculation purpose. The
spatial covariance matrix contains all important information of the mobile radio channel. Further it is
the most flexible way because a lot of different beamforming features like angular diversity or null
broadening can be included in this system matrix with no need to change the used weight calculation
approaches. We investigated two different methods to estimate the spatial covariance matrix at the
downlink frequency.
2.2.2.1 Direction based Approach
Null
Broadening
Weight
Calculation
(Beamforming)
Transmit
Power
Calculation
Herein we first estimate the dominating DOAs (directions-of-arrival) and the corresponding powers
coming from that specific directions. For high-resolution DOA estimation we applied a spatial
reference algorithms called Unitary ESPRIT [HN_95]. Afterwards the estimated DOAs are assigned to
the specific users by utilising user identification algorithms. We allocated the DOAs ideally, because
this was not part of our investigations. With these estimates on the hand you can create the spatial
covariance matrices of the downlink
"Smart Antennas for Mobile Communications Systems" 6

P-12147 MAT Project Report 1997-98
L
H
R = ∑ P ( Θl
) a(
Θl
, f d ) a(
Θl
, f d ) ,
l=
1
where a(Θl,fd) denotes the array steering vector of the direction Θl at the downlink frequency fd. The
power values P(Θl) were estimated using a simple beamformer.
2.2.2.2 Transformation based Approach
In an FDD system the frequency dependent array response vector and thus also the spatial covariance
matrix changes significantly from up- to downlink. The idea of this approach was to estimate the
spatial covariance matrix at the uplink and apply a frequency transformation. We investigated two
already proposed algorithms, which should perform this necessary conversion.
The first approach uses a Compensation Matrix [Zet_97] utilising averaging over the angular domain.
This compensation matrix decreases the error for small frequency shifts and inter-element distances of
the antenna array. But this method changes also the mathematical structure of the spatial covariance
matrix, as shown in [Hug_98].
The Duplex Array Approach [RDJP_95] assumes dominant angles of arrival and small duplex
frequencies. This restricts the application to narrowband systems and scenarios with negligible angular
spread.
Both methods did not lead us to satisfactory results. But especially this transformation will be
important for 3 rd generation systems, where the number of users will exceed the number of available
antenna elements by far and therefore makes the application of DOA based methods impossible.
2.2.3 Null Broadening
Null broadening should provide wide enough nulls in the antenna pattern to reduce the produced
interference as far as possible. Array imperfections, calibration errors and a limited channel estimation
accuracy (spatial covariance matrix or DOA) lead us to the urgency of null broadening. Especially in
suburban and urban environments with large spreading of the user signals over the angular domain
(angular spread AS), the performance loss without null broadening is considerable. As a consequence
null broadening in such propagation environments is necessary. If we use exclusively discrete DOAs
for covariance matrix approximation, the beamforming algorithms produce sharp nulls in the
directions of the interferers, illustrated in Fig. 1.2.2. We demonstrate the effect of null broadening on
the antenna pattern in Fig. 1.2.3.
dB
0
−10
−20
−30
−40
−50
User1
User2
Antenna pattern
−60
−80 −60 −40 −20 0 20 40 60 80
Azimuth in degrees
−60
−80 −60 −40 −20 0 20 40 60 80
Azimuth in degrees
Fig. 4. Antenna pattern without null broadening Fig. 5. Antenna pattern with null broadening
"Smart Antennas for Mobile Communications Systems" 7
dB
0
−10
−20
−30
−40
−50
User1
User2
Antenna pattern

P-12147 MAT Project Report 1997-98
We studied methods to modify the covariance matrix in a way that the used beamforming algorithms
automatically produce broad nulls in the direction of the co-channel users. The analysed null
broadening approaches are called Higher-Order Null Broadening [GNB_97], Angular Spread based
Approach [RGV_97] and Multiple Nulling [Stey_86]. The null broadening related part of our
downlink work is subject of one of our papers [HLB_99], where we show a gain of up to 5dB in SNIR
(Signal-to-Noise-plus-Interference Ratio). The three null broadening schemes have comparable
performance but the angular spread based approach is most flexible.
2.2.4 Beamforming Algorithms
We considered beamforming algorithms based on the spatial covariance matrix. Four methods where
taken from literature and one new beamformer was created.
These algorithms can be divided in two subgroups:
• algorithms with interference suppression
- Summed Inverse Carrier-to-Interference Ratio Minimiser (SICR) [Zet_97]
- Interference Minimisation [FN_95]
- Linearised Power Minimiser (LPM) [Far_97]
- Interference Rejection Method (IRM) [Hu_98] - our beamforming algorithm
• algorithms without interference suppression
- Maximum Desired Power (MDP) [Zet_97].
Our simulations indicated that interference suppression is essential in an SDMA (Space Division
Multiple Access) system. The produced interference with no nulling of the co-channel users is too
large to provide a satisfactory link quality.
Our new algorithm, the IRM (Interference Rejection Method), showed similar results with a
computational complexity smaller by a factor of 3 compared with the best of the other considered
algorithms [Hu_98]. Nevertheless the standard algorithm in the literature, the SICR (Summed Inverse
Carrier-to-Interference Ratio Minimiser), was used in most of the performed simulations because
most of the scientists use this algorithm for weight calculation purpose. Therefore it was possible for
us to compare our downlink performance with already published results.
2.2.5 Downlink Transmit Power
The calculated antenna weights, using beamforming algorithms, define only the shape of the produced
antenna pattern. But it is also essential to balance the transmitter power of the co-channel users to
maximise the SNIR of all users.
We analysed three different transmitter power calculation methods. Transmitting the signals with
equal power from the BS to each user lead to better results than the two other, more sophisticated
approaches (Carrier Balancing and Carrier-to-Interference Balancing [SB_97]). The reason the
impossible prediction of the actual path loss even if the mean power values are already known. Further
the mean path loss of MSs served on the same physical channel varies only slightly, if ambitious
channel allocation schemes are used. This channel allocation provides the spatial separability of the
MSs (the mobiles do no lie in the same direction seen from the base) and the mobiles are arranged in
power classes (mobiles in the same power class have similar transmitter powers and path losses)
[Tan_95]. Thus it is absolutely not necessary to apply sophisticated transmitter power calculation
methods in the downlink beamforming process.
"Smart Antennas for Mobile Communications Systems" 8

P-12147 MAT Project Report 1997-98
2.2.6 Simulations and Results
The downlink beamforming methods described above were used for comparative simulations between
different combinations of the interchangeable blocks of the complete downlink beamforming process
illustrated in Fig. 1.2.1.
The performed simulations lead us to the following conclusions:
• DOA estimation is necessary for downlink beamforming in an SDMA-FDD system if it is not
possible to transform the spatial covariance matrix from uplink to downlink frequency.
• Null broadening is vital in real operating SDMA systems because of the angular spreading,
transceiver imbalances and the limited estimation accuracy.
• The SICR showed the best performance of the investigated beamforming algorithms, slightly
better than our IRM algorithm. Nevertheless the IRM performs also very good but has an
extremely reduced computational complexity.
• Transmitting with the same power to each user gives the best link level performance with
sophisticated channel allocation.
• Angular diversity slightly improves the link quality. The simulated diversity gain would have been
larger, if both scattering clusters had equal mean power values. But this is not the case in our
local/far scattering model due to the higher path loss of the far scattering cluster.
A detailed description of all algorithms and simulation results concerning the downlink can be found
in [Hu_98].
2.3 Receiver Imperfections and Calibration
2.3.1 Introduction
Typically in signal processing simulations the hardware parts of the receiver are assumed to operate
ideally. Especially in array processing applications, in which the whole operation is based on several
independent transceiver trains, the non-idealities of the hardware parts reduce the performance of the
whole adaptive antenna system drastically. In practice the antenna pattern with imperfections differs
essentially from the ideal one, shifting the main beams and nulls from their correct directions. Other
recent research projects, demonstrating the adaptive antenna operation with field trials, have also
reported significance of these issues [SB_98]. The differences in independent hardware components or
drifting because of variations in environmental conditions cause nonlinearities in different receiver
trains. In this work we included an extensive error model into our simulation environment and
analysed the performance impairment with two different receiver structures.
In operating mobile systems the effect of these nonlinearities must be eliminated by using some kind
of calibration method. A widely used method is the so-called switched calibration which means that a
pure oscillator signal is regularly coupled to all input antenna ports and the differences between the
receiver train outputs are used to update the calibration coefficients. This well-known method has been
successfully used in different testbed implementations [TB_97]. However, this technique adds to the
complexity of the expensive hardware part, which naturally is not desirable. In addition to that, the
stability of the calibration sub-system can be problematic in practical operation environments e.g. with
very large temperature variations. Normally the stability does not cause problems in testbed
implementations, because operation times and conditions do not necessarily fully correspond to
situation in real networks.
We considered alternative calibration techniques based on the so-called auto-calibration concept.
With these techniques the desired parameters, like DOAs (direction of arrival), and the error factors
corrupting the antenna signals are estimated in a joint way in the signal processing part of the receiver.
"Smart Antennas for Mobile Communications Systems" 9

P-12147 MAT Project Report 1997-98
Auto-calibration schemes have been successfully used in other array processing applications, like
radar and military systems and we tested their applicability on the mobile radio.
This work is described in details in [Sch_98] and results are partially shown also in [LSB_98],
[LKSB_99a] and [LKSB_99b].
2.3.2 Model of Imperfections
In the error model of our simulation environment we included the following error sources:
• Mutual coupling:
The elements of an antenna array interact with each other, i.e. the current of one element induces
voltages also for neighbouring antennas. We modelled this phenomena by means of mutual
impedances leading to an error matrix, where the non-diagonal elements defined the coupling between
the different antenna elements. Defining these coupling coefficients for arbitrary array geometries is a
non-trivial task leading to numerical procedures, but values for practically interesting array topologies
are available in the literature [Kui_97], [ST_81].
• Phase noise of oscillators:
In the receiver one or more oscillators are needed to downconvert the signal from RF (radio
frequency) to baseband. Ideally, the oscillator output would consist only of one single frequency, but
in reality the energy spreads onto the nearby frequency band as well. This effect, mainly caused by
phase noise, perturbs the downconverted signal causing some undesired random phase modulation. In
our error model we describe the phase noise of the local oscillators (LO) by their spectral power
density. This is a commonly used method to model oscillator noise, because it allows simple spectrum
analyser measurements. The parameter selection of our model was based on the measurements of the
commercial receiver chips [Ben_98].
• Imbalances between receiver trains:
The independent transceiver trains connected to the practical adaptive antenna system can never be
exactly equal, because of component spread. Thus, magnitude and phase imbalances destroy the array
manifold which is the underlying structure for the many signal processing schemes (spatial reference
algorithms). We selected these error factors from the Gaussian distribution using different variances.
• Imbalances between I/Q branches of one transceiver train
Especially when analog downconversion from IF (intermediate frequency) to the complex baseband is
used, imbalances will be present also between I/Q (in-phase/quadrature-phase) channels of one
receiver chain. In case of digital downconversion the imbalance between different branches is less
significant.
• Quantisation noise
The number of quantisation steps of the ADC (analog to digital converter) is an important design
parameter affecting the whole subsequent digital part of the transceiver. Naturally, it is desirable to
minimise the number of quantisation steps still allowing the appropriate operation of the receiver. We
included the effect of the quantisation noise in our simulation chain and defined the lower limit for
A/D conversion levels for both considered receivers.
• Cable length differences:
In practice the antenna elements are coupled to the receiver hardware with some kind of cable. Small
differences of the electrical length of the transmission lines cause signal phase shifts, which were also
taken into account in the error model.
"Smart Antennas for Mobile Communications Systems" 10

P-12147 MAT Project Report 1997-98
2.3.3 Simulations and Results
The error model described above was used for comparative simulations between two different receiver
structures. Chapter 2.4 describes more exactly the first considered receiver utilising semi-blind
estimation technique. This algorithm is not based on the information about the array manifold, but
utilises the known signal properties and initialises the estimates by known bit fields (e.g. training
sequences of GSM system). As a reference for performance comparisons we considered also a more
traditional array processing scheme consisting of DOA estimation, followed by a beamformer and a
standard baseband receiver. For DOA estimation we used Unitary-ESPRIT [HN_95], which is a
numerically advanced subspace based approach. The weight vectors for the beamformer were
calculated using generalised Moore-Penrose pseudo-inverse [Gol_89]. Note, that more advanced
beamforming methods could provide some gain in sense of output SNIR, but for this work more
interesting was to consider the relative performance impairment with different imperfections. Finally
the decision feedback equaliser (DFE) combined with differential detector performed signal detection.
The first receiver structure based on semi-blind estimation combines space-time equalisation, signal
separation and detection. Because of its joint estimation principle, it is reasonable to consider the
simulation results in the form of raw BER (without forward error correction). On the other hand, with
another receiver the processing can be divided into separate parts and thus in addition to the final
detected sequences also the DOA estimator and beamformer outputs can be considered separately. We
used three different performance measures for our spatial reference receiver; in addition to the BER
also distributions of the DOA estimation error and SNIR after beamforming were considered.
In our simulation campaign we found out that the semi-blind estimation was much more robust when
different imperfections were taken into account. It is reasonable that spatial reference algorithms
relying on the known array geometry suffer more from imperfections disturbing array manifold
structure. Correspondingly the semi-blind estimator was insensitive against all errors affecting the
phase of the received signals. This is due to the fact, that phase imperfections can be hidden in the
channel impulse responses during the blind estimation process. The only considered imperfection
somehow affecting the semi-blind receiver was magnitude imbalances between different receiver
trains. However, this effect was also observed only with strong imbalance values exceeding 5dB
(Fig.6). The reason for this behaviour is that singular values needed in subspace estimation part of the
algorithm were disturbed by multiplicative magnitude factors.
BER
10 0
10 −1
10 −2
10 −3
Receiver R2, Gain Imbalances
σ =0dB
g
σ =0.5dB
g
σ =1dB
g
σ =3dB
g
σ =5dB
g
σ =10dB
g
10
0 2 4 6 8 10 12 14 16 18 20
−4
SNR/dB
Fig. 6. Semi-blind technique: Fig. 7. Spatial reference technique:
BER with magnitude imbalances DOA estimation errors with magnitude
imbalances, SNR=20dB
The most significant imperfection affecting the performance of the spatial reference receiver was the
phase- and magnitude imbalances between different receiver trains. Both of these effects increased the
"Smart Antennas for Mobile Communications Systems" 11

P-12147 MAT Project Report 1997-98
number of non-correct DOA estimates. Fig. 7. shows the effect of the magnitude imbalances on the
number of the failed DOA estimates. In this simulation the DOA estimate is considered to have failed
when the error exceeds 7,5°. Correspondingly, performance deterioration can also be seen at the
output of the beamformer. Fig. 8. shows the cumulative distribution function (CDF) of the output
SNIR values for the different magnitude imbalance values with input SNR 20dB.
CDF(SNIR)
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
Receiver R1, Gain Imbalances, SNR=20dB
σ =0dB
g
σ =0.2dB
g
σ =0.5dB
g
σ =1dB
g
σ =2dB
g
σ =5dB
g
σ =10dB
g
0
−30 −20 −10 0 10 20 30 40
SNIR/dB
Fig. 8. Spatial reference technique: Fig. 9. Calibration performance using method
Output SNIR with magnitude imbalances, [BCH_91], magnitude imbalances
SNR=20dB
2.3.4 Calibration Considerations
86
σ =1 dB
g
84
σ =3 dB
g
σ =5 dB
g
82
σ =7 dB
g
σ =9 dB
g
80
0 2 4 6 8 10
SNR/dB
12 14 16 18 20
As already shortly discussed above the imperfections present in the real operating systems require
calibration of these disturbing effects. Auto-calibration techniques estimate jointly the needed
parameters and hardware distortions from the baseband signals. Each of these methods minimises
some kind of cost function using appropriate constraints. In our work we considered the applicability
of the following algorithms, mainly developed for other array processing applications, for the mobile
radio systems:
• Method exploiting the orthogonality of the noise subspace and the Hermitian array steering matrix
[ASO_97].
• Calibration technique utilising the Toeplitz structure of the spatial covariance matrix in case that
sources are uncorrelated and uniform linear array is used [PK_85].
• Technique that makes use of the fact that each perturbed array steering matrix can be constructed
as a linear combination of the eigenvectors spanning the signal subspace [BCH_91]
• Calibration scheme using the properties of the eigenvalue decomposition of the spatial covariance
matrix [WF_96]
• A method based on the MAP (Maximum A Posteriori) estimator of the array perturbations
[Swi_96]
In simulations we created a relative performance measure describing the gain obtained by different
calibration schemes. In addition to the Rayleigh fading signals with different angular spread (AS)
values we considered also single waves arriving at the receiver.
All of the investigated algorithms reduced the imperfection levels. As an example Fig. 9. shows the
relative performance measure for the calibration method [BCH_91] with magnitude imbalances in
non-AS conditions. However, many algorithms might have problems to meet the strict requirements of
the mobile radio systems in practical operation conditions. Reference [SB_98] gives the following
"Smart Antennas for Mobile Communications Systems" 12
E R,g /%
100
98
96
94
92
88
Calibration Method C3

P-12147 MAT Project Report 1997-98
requirements for the accuracy of the calibration; for magnitudes variation less than 0,5 dB and for
phases less than 3°. In [She_97] even more stringent values are given. The problems of autocalibration
techniques are related to the fading nature of the propagation channel and the angular
spread destroying the plane wave assumption. Sometimes deep fades lead to totally failed imperfection
estimates, which can even worsen the situation after calibration. However, the performance
degradation because of totally failed estimated can be prevented by employing some kind of tracking
techniques.
2.4 Robust Algorithms (Blind and Semi-Blind Estimation)
2.4.1 Introduction
As discussed above, the robustness of the algorithms against different hardware imperfections is
important feature in the mobile radio application. Thus, one of the aims of the current project has been
to consider and create new robust signal estimation and detection techniques. We have focussed the
research onto the blind techniques which utilise the known signal properties during the estimation
process. Traditional array signal processing techniques are based mainly either on known array
manifold (spatial reference algorithms, MUSIC [Sch_86], ESPRIT [RPK_86] or its more recent
extensions [HN_95]) or utilisation of the training sequences included in the slot structure of the
systems (temporal reference algorithms, e.g. employing well known recursive least-squares techniques
[Fuhl_97]).
The signals used in mobile communication systems include typically several such rich structural
properties which enable their blind estimation. In our work we have mainly employed the following
known features:
• fixed symbol rate: allows the factorisation of the received data onto the channel response and
signal matrices by means of the special structure of the data matrix.
• finite alphabet property (i.e. the limited number of the modulation symbols): enables together with
the first property solving the FIR-MIMO (Finite Impulse Response, Multiple Input, Multiple
Output) problem.
Other useful properties are: cyclostationarity [TXK_94], constant modulus [VP_96a], spectral selfcoherence
[Agee_90] and higher order statistical properties [Car_91].
Utilising these signal characteristics the channel response matrix which maps the simultaneously
transmitted signals to the received array data samples can be identified without the aid of the known
bit sequences.
In addition to the robustness we see also several other benefits which motivate the use of blind
estimation methods [LSB_98]:
• We are not relying on the known array manifold, therefore neither discrete DOAs (direction-ofarrival)
with limited angular spread nor angular separability are required.
• In our blind estimation case there are no requirements on the synchronisation of the incoming
signals.
• The utilisation of the known signal properties leads to reduced need of overhead information,
which means additional capacity increase.
We have shown that purely blind signal separation and detection is possible in SDMA (space division
multiple access) system where several users are served in the same traffic channel [LB_97]. However,
in all current digital mobile radio standards some known bit fields (training sequences) are available.
Utilising this included information together with the known signal properties leads to the semi-blind
estimation concept. The main idea is to exploit all available information (hidden in the training
sequences and signal properties) and thus improve the estimation accuracy and robustness.
"Smart Antennas for Mobile Communications Systems" 13

P-12147 MAT Project Report 1997-98
Nevertheless we omit the spatial information because it is often disturbed by transceiver imperfections
and angular spread present in mobile radio application. Additionally, we need some kind of known bit
fields to allow user identification when several users are present in the same traffic channel (SDMA
application). Naturally, it is reasonable to use this information already in the signal estimation phase,
not only afterwards when the detected bit sequences already are available [LKSB_99a, LKSB_99b].
Our semi-blind estimation method utilises the joint estimation principle combining three tasks: joint
space-time equalisation, separation and detection of the multiple oversampled co-channel digital
signals. Figure.10 illustrates the principle of the method.
s �
s �
P
x
P
x
Fig. 10. Structure of a semi-blind estimator Fig.11. Semi-blind adaptation
The estimation process consists of two parts. First we estimate the basis of the desired subspace and
after that we project these basis vectors to the finite alphabet (FA) constellation. Between these two
steps we increase the robustness of the estimation by initialising the FA projections with user
identification fields. Figure 11 shows the parts of the entire semi-blind estimation process.
2.4.2 Subspace Estimation (Space-Time Equalisation)
The computationally most expensive part of the algorithm is to estimate the basis vectors spanning the
row space of the specially structured array data matrix. In practice this corresponds to the joint spacetime
equalisation of all incoming signals. Employment of the singular value decomposition (SVD)
leads naturally to the optimal bit error rate performance, but the computational complexity can be
decreased by using adaptive subspace tracking algorithms. With these techniques the subspace
estimate is updated iteratively over the columns of the matrix instead of considering it blockwise as a
whole. We have analysed the performance when the required subspace is estimated using PAST
(Projection Approximation Subspace Tracking) and PASTd (Projection Approximation Subspace
Tracking with deflation) techniques [Yang_95] and compared it to the optimal SVD case. The slight
lack of orthonormality with tracking algorithms is eliminated by performing an additional
orthonormalisation step once after the last update. The tracking methods lead to a slightly worse BER
performance, but correspondingly the computational complexity is reduced maximally by the factor of
70 [LKB_99].
2.4.3 DILSF Algorithm
P
x �
���
�
s �
s �
After the subspace estimation part we have to detect different desired signals by performing leastsquares
projections utilising known finite alphabet constellation. We call our projection algorithm
Decoupled Iterative Least-Squares with Subspace Fitting, DILSF [LB_98]. It combines the ideas of
"Smart Antennas for Mobile Communications Systems" 14

P-12147 MAT Project Report 1997-98
the DWILSP (Decoupled Weighted Iterative Least-Squares with Projections) [Pel_97] and the ILSF
(Iterative Least-Squares with Subspace Fitting) algorithms [VTP_97]. Instead of simultaneous
iteration of all symbol vectors, this approach makes the projections between the FA constellation and
the obtained subspace separately for each user, which has several desirable properties. This projection
approach is computationally very efficient, because during each iteration round only matrix vector
multiplications are required. We continue iterations until convergence and after begin then with the
estimation of the next desired signal. In our simulations we needed typically 2-3 iterations before
convergence.
2.4.4 Simulations and Results
We carried out the simulations using the Geometry-based Stochastic Channel Model (GSCM)
described in the previous section 2.1. and more in detail e.g. in references [MKLH_98] and
[MLKS_98]. Appropriate parameter selection allows modelling of the different propagation
environments. In these simulations we used a parameter set corresponding to the urban environment
with base station antennas above the rooftop level. This parameter selection gives an averaged angular
spread (AS) of about 3 degrees related to each nominal direction-of-arrival (DOA).
The results of our simulations show the raw bit error rates (BER) as a function of the input SNR
(signal to noise ratio). In all simulated cases we assumed two SDMA users and averaged the BER over
both of them. The SNR values were defined by the received mean power values averaged over a large
number of random channel situations.
Figure 12. shows the performance with different number of antenna elements and antenna spacing.
When the element number was decreased, the element spacing was increased correspondingly. This
reduces the correlation between the fading signals received by the independent elements and thus
improves the performance compared to the λ/2 spaced scenario. In this simulation the inter-element
spacing was selected so that the overall length of the array was in the range of 20 . Thus the element
spacing was 20 , 10 and 7 corresponding to the array sizes of M=2, M=3 and M=4 elements,
respectively.
Figure 12 compares the performance with the fixed number of the antenna elements (M=8) when the
required subspace was estimated using SVD, PAST and PASTd algorithms. The tracking approach
BER
10 −1
10 −2
10 −3
Different Element Spacing, DILSF
10
−5 0 5 10 15 20
−4
SNR
M=4, d=7 lambda
M=3, d=10 lambda
M=2, d=20 lambda
−5 0 5 10 15 20
SNR
leads to saturation to a certain BER floor level, but the computational complexity decreased by a
factor of 70.
Fig. 12. BER with different element spacing, Fig. 13. BER comparison between PAST, PASTd
M=2, 3, 4 and SVD, /2 spacing, M=8
"Smart Antennas for Mobile Communications Systems" 15
BER
10 −1
10 −2
10 −3
SVD versus subspace tracking
SVD
PAST
PASTd

P-12147 MAT Project Report 1997-98
When traditional array processing algorithms based on DOA estimation are used, the detection of the
different signals is only possible when are separable in angular domain. Our approach estimates the
unknown channel response matrix, and thus also angularly overlapping signals are separable in case of
different channel responses. To demonstrate this property we created the following modified channel
scenario. Instead of random positioning of the mobiles we placed both users near each other (same
MS-BS distance, DOA difference 2°). Additionally we used the same scattering points for both users
and only local scatterers were present in the simulation. Thus, in this worst case scenario (Fig. 14) all
signal components transmitted by two co-channel users were propagated via the same scattering points
before arriving at the base station array.
Clearly no detection based on DOA estimation can cope with this scenario, but the different channel
response matrices allowed signal separation for our algorithm. The BER performance of this scenario
is shown as the broken line in Fig.15 for /2 array with 6 elements. The solid line shows the situation
in which we added independent far scattering areas for both users, but local contributions propagated
still via the common scattering points. This increased the separability of the channel response matrices
and the performance approached that with the standard channel.
2 o
BS
MS 1
MS 2
Common
Scattering
Points
Fig. 14. Common multipath scenario Fig. 15. BER with common multipath scenario,
/2 spacing, M=6
The performance shown in these figures is very promising especially concerning hardware realisation.
The simulations show that already small number of elements with two times oversampling gives
sufficient performance. Note, however, that increasing the number of the independent samples by
using more array elements improved the performance further.
2.4.5 Testbed Software Implementation
−5 0 5 10 15 20
SNR
The algorithm described above has been implemented also for ����� adaptive antenna testbed
developed in our group.
"Smart Antennas for Mobile Communications Systems" 16
BER
10 −1
10 −2
10 −3
Common Multipath Scenario
Common LS Multipaths (only)
Common LS + Independent FS
Reference, Normal Channel

P-12147 MAT Project Report 1997-98
2.5 References
[Agee_90] Agee. et.al., “Spectral self-coherence restoral: A new approach to blind adaptive signal
extraction using antenna arrays”, Proc. IEEE, vol. 78, pp. 753-767, April 1990
[ASO_97] D. Asztély, A. Swindlehurst, B. Ottersten, “Auto Calibration for Signal Separation with Uniform
Linear Arrays”, Technical Report IR-S3-SB-9718, Department of Signals, Sensors and Systems,
Royal Institute of Technology (KTH), Stockholm, Sweden, 1997
[BCH_91] G.C. Brown, J.H. McElellan, E.J. Holder, “Eigenstructure Approach for Array Processing and
Calibration with General Phase and Gain Perturbations”, Proc. IEEE Int. Conf. on Acoustics,
Speech and Signal Processing (ICASSP´91), pp. 3037-3040, Toronto, Canada, 1991
[Ben_98] K. Beneder, “Impact on System Parameters on DECT Receiver Performance”, Diplomarbeit,
Institut für Nachrichtentechnik und Hochfrequenztechnik (INTHF), Technische Universität
Wien, Austria, 1998
[Car_91] J.F. Cardoso, “Super-symmetric decomposition of the fourth order cumulant tensor. Blind
identification of more sources than sensors.”, Proc. IEEE Int. Conf. on Acoustics, Speech and
Signal Processing (ICASSP´91), Vol.5, pp. 3109-3112, Toronto, Canada, 1991
[Egg_98] P.C.F. Eggers, “Generation of base station DOA distributions by Jacobi transformation of
scattering areas”, Electronic Letters 34, pp. 24-26, 1998
[Far_97] C. Farsakh, “Raummultiplex mit intelligenten Antennen in zellularen Mobilfunksystemen”,
Dissertation, Lehrstuhl für Netzwerktheorie und Schaltungstechnik, TU München, München
1997
[FKB_97] J. Fuhl, A. Kuchar, E. Bonek, “Capacity Increase in Cellular PCS by Smart Antennas”, Proc.
IEEE Vehicular Technology Conference (VTC´97), pp. 1962-1966, Phoenix, Arizona, USA,
May 4-7, 1997
[FMB_98] J. Fuhl, A.F. Molisch, E. Bonek, “A Unified Channel Model for Mobile Radio Systems with
Smart Antennas”, IEE Proc. - Radar, Sonar and Navigation: Special Issue on Antenna Array
Processing Techniques, Vol.145, pp. 32-41, Feb. 1998
[FN_95] C. Farsakh, J.A. Nossek, “Channel Allocation and Downlink Beamforming in an SDMA Mobile
Radio System”, The 6 th Int. Symp. on Personal, Indoor and Mobile Radio Comm. (PIMRC´95),
pp. 687-691, Toronto, Canada 1995
[Fuhl_97] J. Fuhl, “Smart Antennas for Second and Third Generation Mobile Communications Systems”,
Doctoral Dissertation. INTHFT, TU Wien. Vienna, Austria. 1997. 305 p.
[GNB_97] A.B. Gershman, U. Nickel, J.F. Böhme, “Adaptive beamforming algorithms with robustness
against jammer motion”, IEEE Transactions on Signal Processing, Vol. SP_45, pp. 1878-1885,
July 1997
[Gol_89] G. Golub, C.F. Van Loan, Matrix Computations, 2 nd Ed, 1989, 642 p.
[GP_96] D. Gerlach, A. Paulraj, “Base Station Transmitting Antenna Arrays for Multipath Environment”,
Signal Processing 54 (1996), pp. 59-73, 1996
[Häm_97] S. Hämäläinen, et.al, “A Novel Interface between Link and System Level Simulations”, ACTS
Mobile Communication Summit, pp. 314-319, Aalborg, Denmark. October 7-10, 1997
[HLB_99] K. Hugl , J. Laurila, E. Bonek, “Downlink Performance of Adaptive Antennas with Null
Broadening”. accepted for publication in IEEE Vehicular Technology Conference (VTC '99),
Houston, Texas, 16-20 May, 1999
[HN_95] M. Haardt, J.A. Nossek, “Unitary ESPRIT: How to Obtain an Increased Estimation Accuracy
with a Reduced Computational Burden”. IEEE Tr. on Signal Proc., Vol. 43, pp. 1232-1242, May
1995
"Smart Antennas for Mobile Communications Systems" 17

P-12147 MAT Project Report 1997-98
[Hu_98] K. Hugl, “Downlink Beamforming Using Adaptive Antennas”, Diploma thesis, Institut für
Nachrichtentechnik und Hochfrequenztechnik (INTHF), Technische Universität Wien, Wien,
May 1998, 89p.
[Kui_97] M. Kuipers, “Berücksichtigung der Elementverkopplung bei der Synthese
signalsituationsoptimierter mehrelementiger Mobilfunkantennen”, Abschlußbericht zum
gleichnahmigen Projekt, Universität Paderborn, Fachgebiet Nachrichtentechnik, Germany, 1997
[LB_97] J. Laurila, E. Bonek, “SDMA Using Blind Adaptation”, ACTS Mobile Communication Summit,
pp. 314-319, Aalborg, Denmark, October 7-10, 1997
[LB_98] J. Laurila, E. Bonek, "Pseudo-Blind Algorithm for SDMA Application", 8th Virginia Tech.
Symposium on Wireless Personal Communications, p. 23-34, Blacksburg, Virginia, USA, June
10-12, 1998
[LKB_99] J. Laurila, K. Kopsa and E. Bonek, "Semi-Blind Signal Separation and Detection for Smart
Antennas Using Subspace Tracking". submitted to IEEE Signal Processing Advances in Wireless
Communications (SPAWC ’99), Annapolis, Maryland, May 9-12, 1999
[LKSB_99a] J. Laurila, K. Kopsa, R. Schürhuber, and E. Bonek, "Semi-Blind Separation and Detection of Co-
Channel Signals". accepted for publication in IEEE International Conference on
Communications (ICC '99), Vancouver, Canada, June 6-10, 1999
[LKSB_99b] J. Laurila, K. Kopsa, R. Schürhuber and E. Bonek, "Performance Analysis of Semi-Blind Signal
Separation and Detection for Smart Antennas", accepted for publication in IEEE Vehicular
Technology Conference (VTC '99), Houston, Texas, 16-20 May, 1999
[LSB_98] J. Laurila, R. Schürhuber, E. Bonek, "Pseudo-Blind Signal Detection and Separation for Smart
Antennas", Symposium on Intelligent Antenna Technology for Mobile Communications,
Guildford, Surrey, UK, July 9-10,1998
[MKLH_98] A.F. Molisch, A. Kuchar, J. Laurila, K. Hugl, R. Schmalenberger, "Geometry-based directional
model for mobile radio channels - principles and implementation”, submitted to European
Transactions on Telecommunications
[MLK_98] A.F. Molisch, J. Laurila, and A. Kuchar, “Geometry-base stochastic model for mobile radio
channels with directional component”, Proc. 2nd Intelligent Antenna Symp., Univ. Surrey, 9th -
10th July 1998.
[MLKS_98] A.F. Molisch, J. Laurila, A. Kuchar, R. Schmalenberger, "Test scenarios for mobile radio
systems with adaptive antennas", ITG-Fachtafung-Wellenausbreitung bei Funksystemen und
Mikrowellensystemen, Oberpfaffenhofen, Germany, May 11-13, 1998
[Nos_98] J.A. Nossek, “On the Capacity Increase of Cellular Mobile Radio through Smart Antennas”,
MTT-S European Wireless '98, pp. 7-12, Amsterdam, Netherlands, October 8-9, 1999
[Ped_97] K.I. Pedersen, et. al., “Analysis of time, azimuth and doppler dispersion in outdoor radio
channels”, Proc. ACTS Mobile Communication Summit ´97, pp. 308-313, 1997
[Pel_97] P. Pelin, “Spatial Diversity Receivers for Base Station Antenna Arrays”, Licentiate Thesis,
Chalmers University of Technology, Göteborg, Sweden, May 1997, 72 p.
[PK_85] A. Paulraj, T. Kailath, “Direction of Arrival Estimation by Eigenstructure Methods with
Unknown Sensor Gain and Phase”, Proc. IEEE Int. Conf. on Acoustics, Speech and Signal
Processing (ICASSP´85), Vol.2, pp. 640-643, Tampa, Florida, USA, 1985
[RDJP_95] G. Rayleigh, S.N. Diggavi, V.K. Jones, A. Raulraj, “A Blind Adaptive Transmit Antenna
Algorithm for Wireless Communication”, Proc. IEEE International Conference on
Communications (ICC´95), pp. 1494-1499, 1995
[RGV_97] J. Riba, J. Goldberg, G. Vàzquez, “Robust Beamforming for Interference Rejection in Mobile
Communications”, IEEE Transactions on Signal Processing, Vol. 45, No.1, pp. 271-275, January
1997
"Smart Antennas for Mobile Communications Systems" 18

P-12147 MAT Project Report 1997-98
[RPK_86] R. Roy, A. Paulraj, T. Kailath, “ESPRIT - A Subspace Rotation Approach to Estimation of
Parameters of Cisoids in Noise”, IEEE Tr. on Acoustics, Speech and Signal Proc., vol.32, pp.
1340-1342, May 1986
[SB_97] R. Schmalenberger, J.J. Blanz, “A Comparison of two different Algorithms for Multi Antenna
C/I Balancing”, The Second European Personal Mobile Communications Conference
(EPMCC´97), pp. 483-490, Bonn, Germany 1997
[SB_98] C.M. Simmonds, M.A Beach, “Downlink Stability Evaluation of the TSUNAMI(II) Adaptive
Antenna Testbed”, ACTS Mobile Communications Summit, Rhodes, Greece, June 8-11
[Sch_86] R.O. Schmidt, “Multiple Emitter Location and Signal Parameters Estimation”, IEEE Tr. on
Antennas and Prop., vol.34, pp. 276-280, March 1986
[Sch_98] R. Schürhuber, “Receiver Imperfections and Calibration of Adaptive Antennas”, Diplomarbeit,
Institut für Nachrichtentechnik und Hochfrequenztechnik (INTHF), Technische Unversität Wien,
Austria, 1998, 98p.
[She_97] M.W. Shelley, “RF Aspects of Intelligent Base Station Antennas, Symposium on Intelligent
Antenna Technology for Mobile Communications”, Guildford, Surrey, UK, July 9-10,1998
[ST_81] W.L. Stutzman, G.A. Thiele, “Antenna Theory and Design”, John Wiley & Sons, New York,
USA, 1981
[Stey_86] H. Steyskal, “Methods for Null Control and Their Effects on the Radiation Pattern”, IEEE
Transactions on Antennas and Propagation, Vol. 34, No.3, pp.404-409, March 1986
[Swi_96] A. Swindlehurst, “A Maximum a Posteriori Approach to Beamforming in the Presence of
Calibration Errors”, IEEE Workshop on Statistical & Array Processing, pp. 82-85, Corfu,
Greece, 1996
[Tan_95] M. Tangemann, “Near-Far Effects in Adaptive SDMA Systems”, The 6 th Int. Symp. on Personal,
Indoor and Mobile Radio Comm. (PIMRC´95), pp. 1293-1297, Toronto, Canada 1995
[TB_97] G.V. Tsoulos, M.A. Beach, “Space Division Multiple Access (SDMA) Field Trials – Part II”,
IEE Proceedings on Radar, Sonar and Navigation: Special Issue on Antenna Array Processing
Tehcniques, Vol. 145, No. 1, February 1998
[TXK_94] L. Tong, G. Xu, T. Kailath, “Blind Identification and Equalization Based on Second-order
Statistics: A Time Domain Approach”, IEEE Tr. on Information Theory, vol. 40, pp. 340-349,
March 1994
[VP_96a] A.-J. van der Veen, A. Paulraj, “An Analytical Constant Modulus Algorithm”, IEEE Tr. on
Signal Proc., vol.44, pp. 1136-1155, May 1996
[VP_96b] A-J. van der Veen, A. Paulraj, “Singular Value Analysis of Time-Space Equalization in the GSM
Mobile System”, Proc. IEEE Int. Conf. on Acoustics, Speech and Signal Processing (ICASSP
’96), pp. 1073-1076, Atlanta, GA, May 1996
[VTP_97] A-J. van der Veen, S. Talwar, A. Paulraj, “A Subspace Approach to Blind Space-Time Signal
Processing for Wireless Communications Systems”, IEEE Tr. Signal Proc., vol.45, pp. 173-190,
Jan. 1997
[WF_96] A.J. Weiss, B. Frielander, “Almost Blind Steering Vector Estimation Using Second-Order
Moments”, IEEE Transactions on Signal Processing, Vol. 44, No. 4, pp. 1024-1027, 1996
[Yang_95] B. Yang, “Projection Approximation Subspace Tracking”, IEEE Tr. Signal Proc., vol.43, pp. 95-
107, Jan. 1995
[Zet_97] P. Zetterberg, “Mobile Cellular Communications with Base Station Antenna Arrays: Spectrum
Efficiency, Algorithms and Propagation Models”, Dissertation, Royal Institute of Technology,
Stockholm, Sweden, 1997
"Smart Antennas for Mobile Communications Systems" 19

P-12147 MAT Project Report 1997-98
3 PROJECT RELATED PUBLICATIONS
Papers:
[P1] Molisch, A.F., Kuchar, A., Laurila, J., Hugl, K. Schmalenberger, R. "Geometry-based directional
model for mobile radio channels - principles and implementation", submitted to European
Transactions on Telecommunications
[P2] Laurila J., Kopsa K., Schürhuber R., and Bonek E., "Semi-Blind Separation and Detection of Co-
Channel Signals", accepted for publication in International Conference on Communications,
Vancouver, Canada, June 6-10, 1999
[P3] Laurila J., Kopsa K., Schürhuber R. and Bonek E., "Performance Analysis of Semi-Blind Signal
Separation and Detection for Smart Antennas", accepted for publication in IEEE Vehicular
Technology Conference, Houston, Texas, 16-20 May, 1999
[P4] Hugl K., Laurila J., Bonek, E. "Downlink Performance of Adaptive Antennas with Null
Broadening", accepted for publication in IEEE Vehicular Technology Conference, Houston, Texas,
16-20 May, 1999
[P5] Laurila J., Kopsa K. and Bonek E., "Semi-Blind Signal Separation and Detection for Smart
Antennas Using Subspace Tracking", submitted to SPAWC ’99 (Signal Processing Advances in
Wireless Communications)
[P6] Laurila J., Molisch A.F., Bonek, E. "Influence of the scatterer distribution on power delay
profiles and azimuthal power spectra of mobile radio channels". IEEE ISSSTA '98- (5 th International
Symposium on Spread Spectrum Techniques and Applications), Sun City, South Africa, September 2-
4, 1998, p. 267-271
[P7] Laurila J., Bonek E., "Pseudo-Blind Algorithm for SDMA Application". 8th Virginia Tech.
Symposium on Wireless Personal Communications, Blacksburg, Virginia, USA, June 10-12, 1998, p.
23-34
[P8] Molisch A.F., Laurila J., Kuchar A., Schmalenberger R., "Test scenarios for mobile radio
systems with adaptive antennas", ITG-Fachtafung-Wellenausbreitung bei Funksystemen und
Mikrowellensystemen, Oberpfaffenhofen, Germany, May 11-13, 1998
[P9] Molisch A.F., Laurila J., Kuchar A., Schmalenberger R., "Test scenarios for mobile radio
systems with adaptive antennas", First COST 252/259 Joint Workshop, Bradford, UK, April 21-22,
1998, p. 162-170
[P10] Laurila J., Molisch, A.F. "Azimuth Spectra and Delay Considerations of the Channel Model
with Directional Component". ITG-Diskussionsitzung, Modellszenarien für Systeme mit intelligenten
Antennen. Kaiserslautern, Germany. December 5, 1997.
[P11] Laurila J., Bonek E., SDMA Using Blind Adaptation. ACTS Mobile Communication Summit.
Aalborg, Denmark. October 7-10, 1997. p. 314-319
"Smart Antennas for Mobile Communications Systems" 20

P-12147 MAT Project Report 1997-98
Diploma Theses:
[T1] Hugl, K. "Downlink Beamforming Using Adaptive Antennas", Diploma Thesis, Vienna
University of Technology, May 1998, 89p.
[T2] Kopsa, K. "Software Implementation of a Pseudo-Blind Algorithm for Adaptive Antennas",
Diploma Thesis, Vienna University of Technology, September 1998, 73p.
[T3] Schürhuber, R. "Receiver Imperfections and Calibration of Adaptive Antennas", Diploma
Thesis, Vienna University of Technology, October 1998, 98p.
Presentations:
[Pr2] Laurila J., Schürhuber R., Bonek E., "Pseudo-Blind Signal Detection and Separation for Smart
Antennas". Symposium on Intelligent Antenna Technology for Mobile Communications, Guildford,
Surrey, UK, July 9-10,1998
[Pr1] Molisch A.F., Laurila J., Kuchar A., "Geometry-Based Stochastic Model for Mobile Radio
Channels with Directional Component". Symposium on Intelligent Antenna Technology for Mobile
Communications, Guildford, Surrey, UK, July 9-10,1998
"Smart Antennas for Mobile Communications Systems" 21