See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/321006304

Adaptive algorithm based on antenna arrays for radio

communication systems

Article in Serbian Journal of Electrical Engineering · October 2017

DOI: 10.2298/SJEE1703301F

CITATIONS READS

12 151

3 authors, including:

Anna V. Lomakina
Southern Federal University
10 PUBLICATIONS 49 CITATIONS

SEE PROFILE

All content following this page was uploaded by Anna V. Lomakina on 05 December 2017.

The user has requested enhancement of the downloaded file.

SERBIAN JOURNAL OF ELECTRICAL ENGINEERING
Vol. 14, No. 3, October 2017, 301-312
UDC: 004.421.6:621.396.67 DOI: https://doi.org/10.2298/SJEE1703301F

Adaptive Algorithm Based on Antenna

Arrays for Radio Communication Systems
Valentin Fedosov1, Andrey Legin1, Anna Lomakina1
Abstract: Trends in the modern world increasingly lead to the growing popularity
of wireless technologies. This is possible due to the rapid development of mobile
communications, the Internet gaining high popularity, using wireless networks at
enterprises, offices, buildings, etc. It requires advanced network technologies with
high throughput capacity to meet the needs of users. To date, a popular destination
is the development of spatial signal processing techniques allowing to increase
spatial bandwidth of communication channels. The most popular method is spatial
coding MIMO to increase data transmission speed which is carried out due to
several spatial streams emitted by several antennas. Another advantage of this
technology is the bandwidth increase to be achieved without expanding the specified
frequency range. Spatial coding methods are even more attractive due to a limited
frequency resource. Currently, there is an increasing use of wireless communications
(for example, WiFi and WiMAX) in information transmission networks. One of
the main problems of evolving wireless systems is the need to increase bandwidth
and improve the quality of service (reducing the error probability). Bandwidth can
be increased by expanding the bandwidth or increasing the radiated power.
Nevertheless, the application of these methods has some drawbacks, due to the
requirements of biological protection and electromagnetic compatibility, the
increase of power and the expansion of the frequency band is limited. This problem
is especially relevant in mobile (cellular) communication systems and wireless
networks operating in difficult signal propagation conditions. One of the most
effective ways to solve this problem is to use adaptive antenna arrays with weakly
correlated antenna elements. Communication systems using such antennas are
called MIMO systems (Multiple Input Multiple Output multiple input - multiple
outputs). At the moment, existing MIMO-idea implementations do not always
noticeably accelerate traffic at short distances from the access point, but, they are
very effective at long distances. The MIMO principle allows reducing the number
of errors in radio data interchange (BER) without reducing the transmission rate
under conditions of multiple signal re-reflections. The work aims at developing an
adaptive space-time signal algorithm for a wireless data transmission system
designed to improve the efficiency of this system, as well as to study the efficiency
of the algorithm to minimizing the error bit probability and maximizing the
channel capacity.

Keywords: MIMO, OFDM, Channel, Adaptation, Channel capacity.

1
South Federal University, Nekrasovskiy 44, Taganrog 347922, Rostov reg., Russia;
E-mails: [email protected]; [email protected]; [email protected]

301
V. Fedosov, A. Legin, A. Lomakina

1 Introduction
Since the frequency resource is limited and has a restriction in application
in various areas, for example, in mobile communications, wireless local area
networks (Wi-Fi). Also, frequency range allocated for local networks, many
different devices can be sources of interference for other devices. Also, the
reflection of a useful signal from local objects can be perceived as a hindrance.
This leads to the necessity of developing adaptive algorithms to suppress
interference sources improving the quality of data transmission.
Adaptive algorithms can be applied in various fields. Sources of
interference can be of different nature. For example, it could be a source of
deliberate interference to a communication link failure or a source of industrial
interference from equipment at factories, etc. Another example is useful
components of the interfering signal reach the receiving point after multiple
reflections of surrounding objects. These algorithms can be used in
communication devices with the base station and a plurality of mobile stations.
In this case, the mobile devices share a single propagation medium and one
frequency resource, wherein a mobile device of the base station is an active
interference to the other device. In this case, the task of the base station is the
formation of a corresponding radiation pattern. For this kind of tasks, MIMO
technology is also applied and it is called Multi-User MIMO [1].
The use of spatial coding and antenna arrays (AA) on both sides of the
communication system is a promising solution that will significantly increase
spectral efficiency of the allocated frequency resource. Multipath dispersion
except negative impact can be used for data transmission between the
transmitter and the receiver, having no straight path.
Methods based on multiplexing transmission of signals using multiple
antennas, across space. They can be separated by spatial separation (SDM) or
being a form of multiple space division of multiple access (SDMA). These
methods can simultaneously transmit different signals to different, transmitting
antennas, with the same carrier frequency. These parallel data streams interfere
free space but can be reconstructed at the receiver by means of using spatial
signal processing algorithms, which usually require antenna arrays to receive,
and guarantee adequate efficiency concerning the error rate. The difference
between SDM and SDMA is that the latter allows different users to transmit
simultaneously to each single antenna, whereas in a spatial division a user
transmits simultaneously to multiple antennas. Hybrid schemes may also be
proposed. MIMO systems are currently popular due to the lack of frequency
resource, and according to the MIMO method, it is possible to increase
throughput without spreading the signal spectrum. At the same time, the
throughput is increased due to the simultaneous transmission of various data

302
 Adaptive Algorithm Based on Antenna Arrays for Radio Communication Systems

streams through various antenna array elements (AA), and the signals emitted
by each of the antenna array elements occupying a single frequency band.
Using MIMO with other popular technologies and modulation methods
(OFDM, CDMA, COFDM, etc.) it is possible to increase spectral efficiency
significantly, to reduce the bit error and increase the bandwidth. WLAN
standards to focus on, are based on IEEE 802.11a [2] and IEEE 802.11g [3].
The main reason of choosing OFDM technology is the basis of these standards
its noise immunity to the fading channel. Since OFDM allows to reduce, and in
some cases eliminate the effect of fading effect on the signal its application is
popular in digital television systems [4]. In essence, the broadband channel
having frequency selective fading is divided into a plurality of orthogonal
frequency narrow channels with shallow fading (i.e., sub-channels or
subcarriers) from which it can be aligned trivially [5, 6]. This principle is
combined with coding, providing the communication system with stability of
narrowband interference. In addition, it includes the ability to add a proper
guard interval between neighboring OFDM symbols to provide an effective
mechanism dealing with Inter-Character Interference (ISI) caused by
unfavorable multithreaded propagation.
Signals with OFDM are widely used in transmission channels with inter-
symbol interference caused by object reflections. The degree of interfering
action of inter-symbol interference and the probability of error reception
depends on the degree of overlapping of transmitted information symbols.
Therefore, to improve the reception quality of signals under such conditions, it
is advisable to increase the duration of the symbol T. This can be done by
reducing the information transmission rate, which is not always acceptable.
The task of increasing the capacity, as described above, is supposed to be
used in MIMO technology for data transmission. This is possible due to the
ability to separate various spatial streams from multiple transmit antennas,
leading to data rate increase, which depends on the number of antennas of the
transmitter. Also, for the cases with many reflecting environment objects, any
undesirable stream can be minimized or even annihilated in the receiver, and the
stream from the antenna is allocated to the maximum. This separation between
the antennas proves the stability of MIMO systems in a medium with a large
number of signal propagation paths.
To improve the efficiency of such systems, it is necessary to develop an
adaptive algorithm of processing spatio-temporal signals based on antenna
arrays, the essence is consisted in continuous formation of an equivalent
radiation pattern providing interference suppression on the receiving side [7].
This approach will reduce a possible error in wireless data transmission and
send the signal along the path with the greatest power.

303
V. Fedosov, A. Legin, A. Lomakina

2 The Proposed Method

Ideally, if the data rate of MIMO systems grow linearly with the number of
transmitter antennas [8]. However, the maximum transmission rate in a given
frequency band that can be used in MIMO systems depends on some parameters
observed at the receiver, including the average received power of the useful
signal, the thermal noise and the noise associated with the system, as well as
interference from adjacent channels. Also, the multidimensional statistical mode
of the MIMO fading channel is of great significance for the system operation
(for example, the effect of spatial fading correlation). Therefore, it is important
for the designer of MIMO communication system to have an appropriate MIMO
channel modeling design The data of MIMO channel statistics state that radio
signals and system parameters will propagate (such as antenna spacing,
polarization, antenna element orientation). In a real radio propagation
environment, a useful signal can enter the receiving antenna in several ways
with different delay, phase and power. This leads to interference of the received
signal with its copies, and as a consequence it leads to the distortion of the
transmitted information. In this regard, the presence of a mathematical model of
a spatial channel is necessary for the possibility of modeling communication
systems that are close to the real conditions of radio signals propagation. Many
various models differ in implementation complexity. The difficulty consists in a
number of parameters of real environment to be taken into account in a
particular model. Although the implementation of various models is different,
but the result remains the same, it is to obtain the frequency and time
characteristics of the channel in the form of impulse response and amplitude-
frequency. The impulse response reflects all the paths of a signal arrival with
delays and fading, and the frequency features characterizing the channel fading
From the various studies of the channel [9, 10] and many other data given
in the literature, it is supposed that the average multi-path power of the
transmitted pulse in the environment tends is decreased exponentially for a long
period. Also, in general, the amplitudes of the channel coefficient counts are
Rayleigh distributions. In this case the communication channel has a Direct
Visibility (LOS) or a mirror component, as it can be assumed that the first path
is distributed according to Rayleigh-Rice. These features are associated with
such characteristics as:
– slow movement of objects and/or transceivers, leading to a slight Doppler
shift;
– (in the case of LOS) a component that enters the antenna array
completely thus determining from the antenna to the antenna, from the
packet to the main packet;
– a large number of reflectors within a usual similar internal environment
(leading to the Rayleigh fading);
– a feature of propagation loss, where the longer propagation paths of
arriving linearly are later, the energy attenuation is greater.

304
 Adaptive Algorithm Based on Antenna Arrays for Radio Communication Systems

The simplest is the Rayleigh and Rice channel models. These models allow
to imagine an appropriate picture of the signal propagation in a fading channel
[11 – 13]. Reflection from objects leads to the appearance of delayed versions
of the signal in the receiver, the radio signal is subjected to scattering and
attenuation. All these components are combined in the receiver to cause a
phenomenon known as fading. Because of this phenomenon, each path acts as a
discrete fading path. Typically, the attenuation process is characterized by a
Rayleigh distribution if there is no direct visibility between the base station, the
mobile station and the Rice distribution of the case with direct visibility
between the transmitter and the receiver. There are also complex models, for
example based on the 3GPP standard, used in mobile communication systems.
This model of the MIMO channel, proposed by the 3GPP consortium (3rd
Generation Partnership Project) [14], developed for third-generation mobile
communication.

Fig. 1 – Pulse characteristic of 3GPP data transmission channel.

Fig. 2 – Frequency response of 3GPP data transmission channel.

305
V. Fedosov, A. Legin, A. Lomakina

And it is a mathematical model of a wireless distribution channel for dif-

ferent scenarios. Figs. 1 and 2 show the impulse and frequency characteristics
for such model, respectively.

3 Mathematical Model
Multipath propagation of signals is a problem that can cause fading in
communication channel, which is known as a negative factor affecting the
quality of radio communication [9, 12]. Nevertheless, the analysis of the results
of studies conducted by world scientists shows that the multipath for MIMO-
based communication systems may be a favorable factor for wireless
communications in some cases. This is due to the absence of sight line between
the transmitter, and the receiver while data transmission is carried out by the
signal transmission reflected from local objects, for example from buildings.
Also using several independent spatial streams, it is possible to increase the
throughput of simultaneously transmitting data by using several antennas. At
the receiver, the received data symbols are being extracted using various space-
time decoding techniques. To ensure effective separation of different data
streams, AA elements should be spaced apart by a sufficient distance (usually
more than half the carrier wavelength) in order to prevent spatial correlation
between the received signals from different antennas. Fig. 3 illustrates the
MIMO system.
The figure shows the example when a MIMO system is used for the case
with two transmitting antennas, and two receivers. It can be seen from the figure
that the receiving antenna gets a mixture of signals from all antennas and the
task of the receiver is to separate these signals based on the channel state
information.
During the transmission of information on the communication channel from
the base station to the mobile station, the characteristics of the channel, and as a
consequence, the signal parameters can vary due to the following factors: 1) the
presence of the receiver moving relatively to the transmitter, 2) the path of the
received signals, 3) the environment (weather conditions, availability of
jamming sources).

Fig. 3 – MIMO-system.

306
 Adaptive Algorithm Based on Antenna Arrays for Radio Communication Systems

The effect of propagation channel on signal propagation can be

characterized by the transfer characteristic
 h11 (t , ) h12 (t , ) h13 (t , ) ... h1Nt (t , ) 
 
 h21 (t , ) h22 (t , ) h21 (t , ) ... h2 Nt (t , ) 
H (t , )   h31 (t , ) h32 (t , ) h33 (t , ) ... h3 Nt (t , )  . (1)
 
 . . . ... . 
 hN 1 (t , ) hN 2 (t , ) hN 3 (t , ) ... hN r Nt (t , ) 
 r r r

The lines in formula (1) are the transmission factors from all the
transmitting antennas to one of the receivers. The columns show the
transmission factors from one of the transmit antennas to all receivers.
In the MIMO system, the received signal can be described as the following
expression
y  Hx  n , (2)
where H is the channel transmission matrix; y is NR×1 is the vector of the
received signal; x is NT×1 is the vector of the transmitted signal; n is NR×1
receiver noise vector.
The formulas (1) and (2) ensure the matrix form of the received signal
formulas
 y1   h11 h12  h1NT   x1   n1 
y      
 2    h21 h22 h2 NT   x2   n2 
 .
          
      
 y N R   hN R 1 hN R 2  hN R NT   xNT   nN R 
The above written formulas, determine the main task of estimating the
received signal is to estimate the channel matrix.
The task of the adaptive algorithm is to generate a radiation pattern having
a maximum in the direction of arriving useful signal, and the direction of the
interference, a minimum or zero.
Antenna array (AA) receives the signal mixture with interference. With the
help of a constant weight vector v, after the elements of the AP, the formation
of spaced DLs takes place. Then, the received signals are weighted, by means of
the coefficients w, adjusted with the help of an adaptive processor, so that the
source of interference is suppressed. After adaptation, there is a space-time
treatment.

307
V. Fedosov, A. Legin, A. Lomakina

v w1
y1
v w2 +

v wn

AA

v w1
yN
v w2 +

v wn

Adaptive processor

Fig. 4 – Block diagram of the receiving antenna system and adaptation unit.

Let’s consider how the calculation of the weight coefficients in the adaptive
processor is performed in the process of signal reception. To do this it is
necessary to calculate the eigenvalues and eigenvectors of the spatial correlation
matrix
1 n 
R  E{ X  (i ) X T (i )}  
n i 1
X (i ) X T (i ) ,

where X is the vector of the received signal; E{·} is the operation of averaging
the results; * – is the Complex conjugation operation; X T – is the transposition
of the vector X.
In the matrix form, the spatial correlation matrix has the following form
 1 1 1 
 n  x1  x1 x2 
2
x1 xn 
n n
 
1 1 1
R    x2 x1  x22  x2 xn  .
n n n
 
1 x x 1 1 2 
  3 n  x3 x2  n 
x
n n n
To find the optimal weight vectors for the computed spatial correlation
matrix, the approach based on finding the eigenvalues of the matrix R(2) is
applied. This method is widely used to suppress interference and isolate signals
in communication systems using antenna arrays. According to the considered

308
 Adaptive Algorithm Based on Antenna Arrays for Radio Communication Systems

approach, the optimal weight vector is an eigenvector corresponding to the

largest eigenvalue of the correlation matrix R. Thus, the expansion of the
correlation matrix has the following form:
R  VΛV H , (3)
where V is the unitary matrix of eigenvectors and Λ is the diagonal matrix of
the corresponding eigenvalues.
The matrix of eigenvalues has the following form
 1 0  0 
 
0 0  0
Λ ,
    
 
 0 0  n 
where  denotes the value of the eigenvalues of the spatial correlation matrix.
The matrix of eigenvectors is
 v11 v21 vn1 
 
V   v12 v22 vn 2  ,
v v2 n vnn 
 1n
where V is an eigenvector of the matrix R.
X W  X H Wопт .
As a result, at the output of the adaptive processing unit, we get only a
useful signal, free from interference.

3 Experimental Results
As a result of performing the entire procedure of adaptive signal
processing, an array must be formed in the receiver, which a decoded
information sequence of bits is transmitted through the communication channel.
The program implements the method of vertical spatial multiplexing. For
testing, the probability of an erroneous signal reception is calculated.
The SNR is a variable, and, there is an error probability for each specific
value. Counting the errors number is carried out for the given SNR, comparing
the transmitted and received vector of the parcels. In this case, experiment
average takes place. Thus, it is possible to obtain the dependence of the error
reception probability on the value of a certain parameter. Initial data contains:
the number of receiving and transmitting elements, the channel matrix at each
time, the vector of radiated parcels.
To evaluate the effectiveness of the developed algorithm, it is necessary to
conduct a number of the following studies:
309
V. Fedosov, A. Legin, A. Lomakina

1. Determine the probability of bit error occurrence from the input SNR
for a different number of elements in each of the receiving antenna
blocks.
2. Estimate the average throughput through the channel, depending on the
specified SNR for different signal processing cases.
To obtain the correct dependencies of error from SNR, it is necessary to use
averaging over a large number of channel implementations, since most of the
parameters of the channel model are randomly assigned, and it is impossible to
predict channel behavior from case to case.
The initial data is obtained using the existing SCM spatial channel model,
adapted to the task.
Let us analyze the dependence of the bit error probability on the signal-to-
noise ratio after applying the adaptive algorithm for various types of digital
modulation in the MIMO communication system. Analyzing the probability of a
bit error before the adaptation algorithm has no sense since the interference
power exceeds the power of the useful signal greatly. As a result, the useful
signal is destroyed. Fig. 8 shows the dependence of the probability of an error in
the transmitted digital stream on the signal-to-noise ratio in dB.

Fig. 8 – The probability of a bit error depending on the given SNR.

Fig. 9 – The capacity of the data channel depending on the specified SNR.

310
 Adaptive Algorithm Based on Antenna Arrays for Radio Communication Systems

From Figs. 8 and 9 one can judge about the effectiveness of the developed
adaptive algorithm. In all cases, the advantage of using adaptation on the
receiving side is evident.

4 Conclusion
The article presents a synthesized adaptive algorithm for spatial filtering of
signals for a multi-path channel, using output signals from the outputs of the
receiving antenna array to eliminate interference sources leading to disruption
of the communication system. The algorithm is based on a linear transformation
of the received signal, is based on linear algebra methods for calculating the
antenna block weights. A block diagram of the communication system is
presented using the developed algorithm.
Despite the fact that the MIMO system itself is capable of evaluating the
channel characteristics and attempting to suppress interference components,
adaptive beam formation is more efficient, especially when it comes to sources
of a high radiation power compared to a useful signal. Using antenna arrays
from non-direction elements, it becomes possible to apply modern methods of
forming digital radiation pattern to isolate the direction of a useful signal
arrival, while all other sources will be suppressed.

5 Acknowledgement
The reported study was supported by the Ministry of Education of the
Russian Federation, research projects № 8.5591.2017/8.9.

6 References
[1] P. Hammarberg, F. Rusek, O. Edfors: Iterative Receivers with Channel Estimation for
Multi-user MIMO-OFDM: Complexity and Performance, EURASIP Journal on Wireless
Communications and Networking, Vol. 2012, No. 1, Dec. 2012, p. 75.
[2] A. Doufexi, S. Armour, M. Butler, A. Nix, D. Bull, J. McGeehan, P. Karlsson: A
Comparison of the HIPERLAN/2 and IEEE 802.11a Wireless LAN Standards, IEEE
Communications Magazine, Vol. 40, No. 5, May 2002, pp. 172 – 180.
[3] D. Vassis, G. Kormentzas, A. Rouskas, I. Maglogiannis: The IEEE 802.11g Standard for
High Data Rate WLANs, IEEE Network, Vol. 19, No. 3, May/June 2005, pp. 21 – 26.
[4] G. Acosta: OFDM Simulation using Matlab, Aug. 2000.
[5] A. Van Zelst: MIMO OFDM for Wireless LANs, PhD Thesis, Eindhoven University of
Technology, Eindhoven, Netherlands, 2004.
[6] M. Ibnkahla: Handbook on Signal Processing for Mobile Communications, CRC Press,
BocaRaton, FL, USA, 2004.
[7] V.P. Fedosov, A.V. Emelyanenko, N.O. Ternovay, R.V. Rubtsov, V.I. Marchuk: Adaptive
Algorithm for Mimo-System of the Wireless Access for the Receiver of the Mobile Station,
12th International Conference on Signal Processing Proceedings, Hangzhou, China, 19-23
Oct. 2014, pp. 1537 – 1541.

311
 V. Fedosov, A. Legin, A. Lomakina

[8] A. Goldsmith, S.A. Jafar, N. Jindal, S. Vishwanath: Capacity Limits of MIMO Channels,
IEEE Journal on Selected Areas in Communications, Vol. 21, No. 5, June 2003, pp. 684 –
702.
[9] E.L.M. Morales, O.E.R. Ramirez, L. De Haro Arriet: Antenna and Wireless Multipath
Virtual Channel Interaction, Asia-Pacific Microwave Conference, Bangkok, Thailand, 11-14
Dec. 2007.
[10] K. Yu, B. Ottersten: Models for MIMO Propagation Channels: A Review, Wireless
Communications and Mobile Computing, Vol. 2, No. 7, Nov. 2002, pp. 653 – 666.
[11] M. Chiani, M.Z. Win, A. Zanella: On the Capacity of Spatially Correlated MIMO Rayleigh-
fading Channels, IEEE Transactions on Information Theory, Vol. 49, No. 10, Oct. 2003, pp.
2363 – 2371.
[12] A.A.M. Saleh, R. Valenzuela: A Statistical Model for Indoor Multipath Propagation, IEEE
Journal on Selected Areas in Communications, Vol. 5, No. 2, Feb. 1987, pp. 128 – 137.
[13] D.S. Shiu, G.J. Foschini, M.J. Gans, J.M. Kahn: Fading Correlation and its Effect on the
Capacity of Multielement Antenna Systems, IEEE Transactions on Communications, Vol.
48, No. 3, March 2000, pp. 502 – 513.
[14] Physical Layer Aspects for Evolved UTRA, 3GPP TR 25.814 V7.1.0, Sept. 2006. Available
at: http://www.3gpp.org/ftp/Specs/html-info/25814.htm.

312

View publication stats