<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Archiving and Interchange DTD v1.0 20120330//EN" "JATS-archivearticle1.dtd">
<article xmlns:xlink="http://www.w3.org/1999/xlink">
  <front>
    <journal-meta>
      <journal-title-group>
        <journal-title>International Journal of Computer Network and Information Security 16 (2024) 31-51.
doi:10.5815/ijcnis.2024.04.03.
[28] O. Shcherbyna</journal-title>
      </journal-title-group>
    </journal-meta>
    <article-meta>
      <article-id pub-id-type="doi">10.1109/8.884492</article-id>
      <title-group>
        <article-title>Selection of antenna elements for a radio emission source detection and tracking system</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <string-name>Olga Shcherbyna</string-name>
          <email>olha.shcherbyna@npp.kai.edu.ua</email>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Fedir Katushonok</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Oleksandr Zadorozhnyi</string-name>
          <email>oleksandr.zadorozhnyi@npp.kai.edu.ua</email>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Myroslav Riabyi</string-name>
          <email>myroslav.riabyi@npp.kai.edu.ua</email>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <aff id="aff0">
          <label>0</label>
          <institution>State University “Kyiv Aviation Institute”</institution>
          ,
          <addr-line>Liubomyra Huzara Ave., 1, Kyiv, 03058</addr-line>
          ,
          <country country="UA">Ukraine</country>
        </aff>
      </contrib-group>
      <pub-date>
        <year>2019</year>
      </pub-date>
      <volume>1418</volume>
      <fpage>1235</fpage>
      <lpage>1245</lpage>
      <abstract>
        <p>In the presented article, a review was conducted on the existing antenna designs that can serve as prototypes for the antenna segment of detection and tracking systems for radio emission sources (RES). Selecting the proper design of the antenna segment for such systems is one of the most critical steps in their construction. For amplitude-based direction-finding methods, two types of directional antennas were modeled and analyzed: the Yagi-Uda antenna and the patch antenna array 4x2. The patch antenna array 4x2 ofers greater accuracy in determining the location of the RES due to the formation of a narrower radiation pattern and significantly wider frequency bandwidth, enabling the direction-finding system to operate without the need to replace the antenna unit. However, there are also drawbacks: the presence of relatively high-level side lobes (up to -10 dB) in the radiation pattern, which may lead to ambiguity in determining the RES direction, as well as significantly larger dimensions compared to the Yagi-Uda antenna. For implementing a phase-based method of determining the RES position, a 2x2 antenna array was used, consisting of half-wave dipole antennas with precisely defined phase centers. To eliminate ambiguity in detecting the RES direction, the dipole array is positioned above a reflective surface and includes a dual-position rotating mechanism.</p>
      </abstract>
      <kwd-group>
        <kwd>eol&gt;antennas</kwd>
        <kwd>Yagi-Uda antenna</kwd>
        <kwd>antenna array</kwd>
        <kwd>direction finding</kwd>
        <kwd>radio emission source</kwd>
        <kwd>amplitude method</kwd>
        <kwd>phase method</kwd>
      </kwd-group>
    </article-meta>
  </front>
  <body>
    <sec id="sec-1">
      <title>1. Introduction</title>
      <p>Accurately determining the position of radio emission sources (RES) in space is a critical task in various
ifelds, ranging from civilian (governmental, scientific, and commercial) to military applications. The
position of RES can be static or dynamic, single or multi-positional, and either friendly or hostile.
Determining the position of RES involves taking into account parameters such as spatial angles, the
distance to the object, and the object’s velocity.</p>
      <p>
        The function for detecting and tracking RES, often referred to as the direction-finding function, is
one of the most critical components of modern radio systems and complexes for detection, ranging,
monitoring, navigation, control, and communication [
        <xref ref-type="bibr" rid="ref1 ref2 ref3">1, 2, 3</xref>
        ].
      </p>
      <p>
        The performance of these systems directly depends on the quality of the direction-finding segment.
Among the key characteristics of such systems are the operating frequency range, direction-finding
accuracy, and the angular range for unambiguous measurements. These parameters are subject to
particularly stringent requirements, especially when direction-finding systems are used for unmanned
aerial vehicles (UAVs) [
        <xref ref-type="bibr" rid="ref4 ref5">4, 5</xref>
        ]. Modern trends in the expansion of the use of UAVs, unmanned maritime
vessels, and ground vehicles [
        <xref ref-type="bibr" rid="ref6 ref7 ref8">6, 7, 8</xref>
        ] necessitate the enhancement of onboard and ground-based systems
for identifying and tracking the angular position of radio emission sources. The primary challenge in
the operation of direction-finding devices lies in balancing high direction-finding accuracy with a wide
angular measurement range.
      </p>
      <p>A promising direction in the development of systems for detecting and tracking radio emission
sources is the integration of Software Defined Radio (SDR) as a system component. SDR primarily relies
on software implementations for most physical functions, while hardware functions are dynamically
adapted to meet the requirements of the communication standard in use. Due to the software-based
execution of high-frequency signal processing functions and adaptive software control, SDR delivers
significant improvements in functionality by operating across a wide frequency range and accommodating
various communication standards.</p>
      <p>
        SDR performs the functions of a radio receiver and transmitter either through software configurations
or programmatically controlled hardware components. Components that cannot be software-defined,
such as power amplifiers or antennas, are still managed within this framework. SDR devices are
celebrated for their flexible configuration, accessibility, and speed, making them widely applied in various
tasks related to communication signal research and processing, including studies of communication
networks, telemetry programs, and applications for measuring the Angle of Arrival (AoA) [
        <xref ref-type="bibr" rid="ref10 ref11 ref9">9, 10, 11</xref>
        ].
      </p>
      <p>
        Numerous studies have explored various methods for measuring AoA. These approaches include the
combined amplitude-phase method [
        <xref ref-type="bibr" rid="ref12">12</xref>
        ], the Multiple Signal Classification (MUSIC) method [ 13], the
beam-switching method [14] relying on simple cross-correlation, and the time-reversal method designed
for detecting low-angle targets [15]. Typically, these methods require the use of complex antenna arrays
and involve substantial processing time. Another approach to AoA measurement involves calculating
the phase diference between two antennas. Additionally, it is important to ensure the separation of the
target signal from any interference signals. To achieve this, researchers commonly employ techniques
such as Independent Component Analysis (ICA) [16] and frequency filtering.
      </p>
    </sec>
    <sec id="sec-2">
      <title>2. Problem statement</title>
      <p>The selection of an appropriate antenna segment design is among the most critical steps in constructing
any radio system, including systems for identifying and tracking RES. Depending on the chosen method,
these systems typically employ highly directional antennas with high gain coeficients or arrays of
omnidirectional antennas arranged according to a specific algorithm. For instance, to implement the
direction-finding function using the amplitude method, it is possible to utilize directional antennas
such as spiral, parabolic, horn antennas, Yagi-Uda antennas, log-periodic antennas, or linear and planar
arrays.</p>
      <p>The use of passive antenna arrays (AAs) facilitates the creation of radiation patterns with narrow
main lobes or patterns of specialized shapes, increased radiation power, enhanced technological design,
and more. AAs allow the combination of multiple weakly directional radiating elements to establish
the desired field radiation distribution in space, implement aperture control for the quality of radiation
pattern formation, enable adjustments, and simultaneously generate multiple radiation patterns.</p>
      <p>Numerous studies have focused on the principles of design, modeling, and analysis of AA with
various configurations. The work [ 17] presents a simple and compact Yagi-Uda antenna constructed
on Rogers RO4003 substrate for multi-band radar applications. This prototype operates within three
distinct frequency bands, namely 1.9 GHz, 2.5 GHz, and 3.5 GHz, achieving gain coeficients of 6.29 dBi,
4.63 dBi, and 6.77 dBi respectively.</p>
      <p>The integration of spiral elements of varying lengths enables the realization of a dual-band spiral
antenna, as explored for frequencies of 1.227 GHz and 1.575 GHz in study [18]. For these frequencies,
gain coeficients of 4.44 dBi and 3.87 dBi were achieved, with a Half-Power Beam Width (HPBW)
values of 96 degrees and 122 degrees, respectively. The principles behind the construction of wire and
microstrip four-element spiral antennas are outlined in studies [19, 20]. Modeling was conducted for
variations of wire and printed quadrifilar spiral antennas designed for a frequency of 1268.52 ± 12 MHz.
The modeling demonstrated a gain coeficient of approximately 3 dBi with an HPBW of 90 degrees.</p>
      <p>In study [21], a comparative analysis was conducted on the characteristics of microstrip log-periodic
antennas (LPAs) within the frequency range of 2-6 GHz using dielectric substrates with varying
properties. Following both modeling and experimental investigation of three prototypes, a gain of over
6 dBi across the entire frequency band was achieved. Additionally, research [22] presented an LPA
design suitable for WLAN/LTE/UWB applications, demonstrating a frequency range extending from
1.4 GHz to 12 GHz and a gain of 4.51 dBi.</p>
      <p>In article [23], a design for a microstrip AA 2x2 with rectangular patch elements for a frequency
of 3.8 GHz is presented, achieving HPBW of 33 degrees and a gain coeficient of 13.2 dBi. Study [ 24]
proposes a compact two-element patch AA with vertical polarization within the frequency range of 2.4
GHz, which achieved a gain coeficient of 9.14 dBi with an HPBW of 60 degrees in the E-plane and 65
degrees in the H-plane during modeling and investigation.</p>
      <p>To implement the phase-based direction-finding algorithm, linear or planar AAs are recommended
as antenna segments. These arrays consist of relatively simple antennas with clearly defined phase
centers, such as half-wave dipoles.</p>
      <p>Furthermore, integrating antenna elements with SDR technology facilitates the development of
antenna systems with digital signal processing capabilities. These systems can efectively analyze
the interference environment, manage antenna radiation patterns for improved user interaction, and
suppress undesirable sources of interference simultaneously. Modeling the characteristics of antennas
utilized for locating and tracking RES using amplitude and phase methods was performed within the
Altair Feko 2024.0.1 software environment. The measurement of antenna matching characteristics was
conducted using a portable vector network analyzer LiteVNA 64.</p>
    </sec>
    <sec id="sec-3">
      <title>3. Antenna elements of detection and tracking systems for RES based on the amplitude method</title>
      <p>In general, the detection and tracking of RES are implemented using two primary methods [25]: the
amplitude method and the phase method. Amplitude-based methods rely on the analysis of the amplitude
distribution of the electromagnetic field generated by the RES within the aperture of the
directionifnding antenna across a specific plane, either azimuthal or meridional. Amplitude direction-finding can
be conducted through various approaches, such as rotating a single high-gain antenna [26], switching
between multiple high-gain antennas, or electronically cycling through low-gain antennas arranged in
a circular array to simulate the rotation of a narrow beam.</p>
      <p>There are three principal amplitude-based direction-finding techniques: the maximum amplitude
method, the minimum amplitude method, and the equal-signal method.</p>
      <p>When using the maximum amplitude method, the position of the RES is determined by identifying
the maximum of the radiation pattern (RP) of the rotating antenna system (Figure 1, a). In this case, the
voltage  of the antenna output directly correlates with the E-field intensity in its plane and the RP of
the antenna  (, ) , where  and  represent the meridional and azimuthal coordinates, respectively:
 = ℎ (, ),
where ℎ is efective height of the direction finder antenna.</p>
      <p>The determination of angular coordinates for RES occurs at the moment when the voltage level at
the output of the radio receiver (RR) reaches its maximum value.</p>
      <p>Advantages of the method:
- minimal influence of noise on detection accuracy, as the maximum RP ensures the highest possible
level of the desired signal is received;
- relative simplicity in implementation, requiring only a single-channel RR.</p>
      <p>Disadvantages of the method:
- insuficient measurement precision due to the low gradient of the antenna’s RP in the vicinity of its
maximum, necessitating antennas with a narrow main lobe in the RP;
- potential errors caused by the presence of side lobes in the RP.</p>
      <p>The accuracy of RES positioning is approximately:
where  0.5 is the HPBW.</p>
      <p>The method of RES detection based on the minimum amplitude of the signal (the diferential method)
is utilized when it is possible to form the antenna system’s RP with a distinct minimum of the received
signal, which must also be singular. Such a RP can be established using a system comprising two
identical narrow-directional antennas (Figure 1, b). The detection of RES is conducted by rotating the
antenna system until the position of the minimum voltage value at the output of the RR is achieved.</p>
      <p>Advantages of the method:
- greater accuracy in determining the coordinates of RES compared to the method based on the
maximum signal, as the RP of the antenna exhibits a steeper gradient near its minimum value.</p>
      <p>The error margin is
Δ = 0.1
0.5.</p>
      <p>(3)
Disadvantages of the method:
- due to the low level of the diferential signal, detection errors can be influenced by noise;
- the operational range is shorter compared to the previous method.</p>
      <p>The equisignal method is designed to eliminate the shortcomings of the first two amplitude-based
methods. It can be implemented through the use of two identical directional antennas, whose RP
are rotated relative to one another by a certain angle  , which is less than the RP width (Fig. 1, c).
The direction towards the RES is determined by achieving equality of the signals received from both
antennas.</p>
      <p>Advantages of the method:
- greater accuracy and sensitivity in determining the coordinates of the RES compared to the minimum
signal method.</p>
      <p>The error margin is
Δ = 0.05
0.5.</p>
      <p>(4)</p>
      <p>As noted above, the implementation of amplitude-based methods for detecting RES frequently
involves the use of one or more antennas with a high gain. For the purposes of this study, two types of
directional antennas were selected: the Yagi-Uda antenna (Figure 2), featuring linear polarization, and
the patch antenna array (Figure 3), characterized by circular polarization. Other types of directional
antennas or antenna arrays [27, 28] may also be employed, depending on the specific operational
conditions. The research was conducted at a central frequency of 5.7 GHz.</p>
      <p>The main radiation and matching characteristics of the studied antennas are shown in Figures 4–9.</p>
      <p>As indicated earlier, the accuracy of determining the location of RES using the maximum signal
method is approximately equation (2), where  0.5 is represents the HPBW in the azimuthal or meridional
plane. It is possible to theoretically calculate the potential error in determining the angular position of
RES when utilizing either a Yagi-Uda antenna or a 4x2 patch antenna array:</p>
      <p>- for the Yagi-Uda antenna, when installed with horizontal polarization:
- for the 4x2 patch antenna array:
in the azimuthal plane:
in the meridional plane:</p>
      <p>0.5 = 0.2 · 57/2 ≈ 6;
Δ = 0.2 0.5 = 0.2 · 76/2 ≈ 8;
Δ = 0.2</p>
      <p>0.5 = 0.2 · 13/2 ≈ 1.3;
Δ = 0.2 0.5 = 0.2 · 31/2 ≈ 3.</p>
      <p>It can be observed that theoretically, the use of a planar antenna array ofers greater accuracy in
determining the position of RES due to the generation of a narrower beam. However, its beam pattern
includes sidelobes (at angles ±27.2 degrees in the azimuthal plane and ±60 degrees in the meridional
plane) with relatively high levels (–10 dB and –15 dB respectively), which may lead to ambiguities in
identifying the exact location of RES.</p>
    </sec>
    <sec id="sec-4">
      <title>4. Antenna elements of the detection and tracking system for RES utilizing the phase methodr</title>
      <p>Phase-based methods for the detection and tracking of RES are founded on the analysis of phase
variations or distributions of signals received by the elements of antenna systems. Phase-based
directionifnding [ 29] employs methodologies such as direct measurement of phase diferences between receivers
equipped with antennas positioned less than half a wavelength apart, interferometric measurement
of phase diferences across multiple spatially separated receivers, and the measurement of signal
phases with phase modulation induced by the Doppler efect due to circular antenna rotation. These
phase measurements in radio systems necessitate post-processing of the received signals. Despite the
diversity of nomenclatures, including Doppler, quasi-Doppler, correlation-phase, interferometric, and
correlation-interferometric methods, all fall under the umbrella of phase-based techniques.</p>
      <p>The phase-based approach to the detection finding and tracking of RES is predicated on utilizing the
dependency of phase diferences in signals received by individual elements of an antenna array, which
are separated by a predetermined distance d (the baseline of the direction finder). This approach can be
exemplified by analyzing a two-element antenna system comprising elements A1 and A2 (Figure 10).</p>
      <p>In general, the electromagnetic wavefront approaching at a certain angle relative to the baseline
degrees (dashed line) and in the plane  = 0 degrees (solid line).</p>
      <p>From this relationship, the angular position can be calculated using the formula:</p>
      <p>Δ =  sin .
Δ =
2Δ

=
2

sin .
between two antennas reaches one antenna earlier than the other. This results in a variation in the
path lengths traveled by the waves arriving at antennas A1 and A2 from a direction determined by the
angle . The diference in path lengths is expressed as:</p>
      <p>Correspondingly, the phase shift observed in the high-frequency oscillations of the signals received
by antennas A1 and A2 is given by:
Accurate determination of the direction to the RES requires knowledge of the frequency (or wavelength)
(5)
(6)
of the received signals and the phase diference of these signals at the reception points.
 = arcsin
Δ
2
.</p>
      <p>(7)
The classical phase method for direction-finding faces a trade-of between the accuracy of measurements
and the unambiguous identification of angular coordinates. The sensitivity of the phase method to
variations in the angle  increases with the relative size of the baseline . However, as the baseline
expands, the angular coordinate value  diminishes, where the phase diference Δ exceeds 2 , leading
to ambiguity in the reading.</p>
      <p>Figure 11 presents the dependencies of angular position determination for RES based on phase
diferences of high-frequency oscillations (7) in the reception circuit, at various relative baseline sizes
of the direction finder.</p>
      <p>If the antenna array system consists of two elements, it may lead to ambiguity in determining the
direction of the signal source since the same phase diference at the antenna outputs can be produced
by a signal arriving from the opposite direction. To eliminate this uncertainty, a second pair of antennas
with a baseline perpendicular to the baseline of the first pair can be employed. Alternatively, shielding
can be utilized. In this case, direction finding is possible only within one half-space, necessitating the
use of a rotational mechanism.</p>
      <p>As mentioned earlier, to implement the phase method for determining the position of the signal source,
it is necessary to use at least two antennas for a two-channel radio receiver. For this study, a 2x2 antenna
array was selected, consisting of half-wave dipole antennas (Figure 12). These antennas are positioned
above a shield to avoid ambiguity in detecting the direction of the signal source. For measuring the
azimuthal angle of the signal source direction, the dipoles are arranged along the horizontal axis,
whereas for measuring the meridional angle, they are arranged along the vertical axis. The frequency
of the signal source is 5.8 GHz. The primary radiation and matching characteristics of the examined
antenna array, when the signal source is located along the line perpendicular to the array plane ( = 0
and  = 0), are illustrated in Figures 13–15.</p>
      <p>However, as the signal source deviates from the line perpendicular to the plane of the antenna array,
the influence of the shield on the accuracy of phase shift determination, and consequently on the
angular position of the bearing, increases. For instance, to illustrate changes in the direction of the
radiation pattern’s maximum (Figure 16) and the impact of the shield as the signal source moves, the
azimuthal direction could be considered.</p>
      <p>As can be observed from a comparison of the graphs illustrating the dependencies of the radiation
pattern’s maximum on the phase shift of feeding elements of the antenna array (Fig. 16), obtained
through simulation, and the calculated values using formula (7), errors Δ arise in determining the
angular position (Table 1). These discrepancies arise due to the influence of the shielding and the
close proximity of adjacent elements on the phase relationships of currents within the antenna array
elements. This highlights the necessity of employing a rotational mechanism to enhance the accuracy
of determining the angle of arrival for signals from RES.</p>
    </sec>
    <sec id="sec-5">
      <title>5. Conclusions</title>
      <p>The selection of an appropriate antenna segment design is one of the critical steps in constructing a
system for detection and tracking of RES. This study presents a review of existing designs that may
serve as prototypes for antennas in tracking systems.</p>
      <p>To implement the amplitude-based direction-finding method, two types of directional antennas were
calculated, modeled, fabricated, and researched: the Yagi-Uda antenna and the patch AA 4x2. Both
prototypes demonstrated excellent radiation and matching characteristics during both simulation and
experimental phases. The gain of the five-element Yagi-Uda antenna within the frequency range of
5.5–5.8 GHz was approximately 9 dBi, with HPBW values of 57 and 76 degrees in the E and H planes,
respectively. The patch AA 4x2 within the frequency range of 5.3–6.3 GHz exhibited the following
parameters: a gain of approximately 17 dBi, with HPBW values of 13 and 31 degrees in horizontal and
vertical planes, respectively. Thus, the use of the patch AA 4x2 ensures greater accuracy in determining
the RES position by forming a narrower beamwidth. Additionally, the patch array has a considerably
broader operational frequency range, allowing signal detection and direction-finding across the entire
bandwidth without antenna block replacement. However, its application has two drawbacks. The first
is the presence of relatively high side lobes in the radiation pattern (up to -10 dB), which may cause
ambiguity in determining the RES position. The second drawback of the patch AA 4x2 is its significantly
larger size compared to the Yagi-Uda antenna (20x10 cm versus 5x3 cm, respectively)</p>
      <p>To implement the phase-based method for determining RES positions, a 2x2 array composed of
halfwave dipole antennas was chosen. These dipole antennas are positioned above a screen to eliminate
directional ambiguity. The dipole array within the frequency range of 5.3–5.9 GHz demonstrated
the following parameters: a gain of approximately 10 dBi and HPBW values of 57 degrees in both
horizontal and vertical planes. However, it was proven that, in the absence of a rotational mechanism,
the movement of the RES from the perpendicular line to the array plane increases the influence of the
screen on the accuracy of determining the phase shift of currents within the antenna elements, thereby
afecting the angular position of direction-finding. Based on the analysis of these results, it was decided
to combine two direction-finding methods – amplitude and phase-based – for the further development
and study of RES detection and tracking systems. These methods will be implemented via an SDR
receiver alongside a combination of the Yagi-Uda antenna and the dipole AA 2x2. The antenna block
will incorporate a rotational device in azimuthal and meridional planes.</p>
    </sec>
    <sec id="sec-6">
      <title>Acknowledgments</title>
      <p>Thanks to the developers of Altair Engineering for the opportunity to use the software product for
calculating electromagnetic environments FEKO when modeling various antenna designs.</p>
    </sec>
    <sec id="sec-7">
      <title>Declaration on Generative AI</title>
      <p>The authors have not employed any Generative AI tools.
2003.1238760.</p>
    </sec>
  </body>
  <back>
    <ref-list>
      <ref id="ref1">
        <mixed-citation>
          [1]
          <string-name>
            <given-names>S.</given-names>
            <surname>Zhyla</surname>
          </string-name>
          , E.Tserne,
          <string-name>
            <given-names>Y.</given-names>
            <surname>Volkov</surname>
          </string-name>
          ,
          <string-name>
            <given-names>S.</given-names>
            <surname>Shevchuk</surname>
          </string-name>
          ,
          <string-name>
            <given-names>O.</given-names>
            <surname>Gribsky</surname>
          </string-name>
          ,
          <string-name>
            <given-names>D.</given-names>
            <surname>Vlasenko</surname>
          </string-name>
          ,
          <string-name>
            <given-names>V.</given-names>
            <surname>Kosharskyi</surname>
          </string-name>
          ,
          <string-name>
            <given-names>D.</given-names>
            <surname>Kovalchuk</surname>
          </string-name>
          ,
          <article-title>Statistical synthesis and analysis of functionally deterministic signal processing techniques for multi-antenna direction finder operation</article-title>
          ,
          <source>Computation</source>
          <volume>12</volume>
          (
          <year>2024</year>
          )
          <article-title>170</article-title>
          . doi:
          <volume>10</volume>
          .3390/ computation12090170.
        </mixed-citation>
      </ref>
      <ref id="ref2">
        <mixed-citation>
          [2]
          <string-name>
            <given-names>A.</given-names>
            <surname>Martian</surname>
          </string-name>
          ,
          <string-name>
            <given-names>C.</given-names>
            <surname>Paleacu</surname>
          </string-name>
          , I. Marcu,
          <string-name>
            <given-names>C.</given-names>
            <surname>Vladeanu</surname>
          </string-name>
          ,
          <article-title>Direction-finding for unmanned aerial vehicles using radio frequency methods</article-title>
          ,
          <source>Measurement</source>
          <volume>235</volume>
          (
          <year>2024</year>
          )
          <article-title>114883</article-title>
          . doi:
          <volume>10</volume>
          .1016/j.measurement.
          <year>2024</year>
          .
          <volume>114883</volume>
          .
        </mixed-citation>
      </ref>
      <ref id="ref3">
        <mixed-citation>
          [3]
          <string-name>
            <given-names>M.</given-names>
            <surname>Zaliskyi</surname>
          </string-name>
          , et al.,
          <article-title>Heteroskedasticity analysis during operational data processing of radio electronic systems</article-title>
          , in: S. Shukla,
          <string-name>
            <given-names>A.</given-names>
            <surname>Unal</surname>
          </string-name>
          ,
          <string-name>
            <given-names>J. V.</given-names>
            <surname>Kureethara</surname>
          </string-name>
          ,
          <string-name>
            <given-names>D.</given-names>
            <surname>Mishra</surname>
          </string-name>
          , D. Han (Eds.),
          <source>Data Science and Security</source>
          , volume
          <volume>290</volume>
          <source>of Lecture Notes in Networks and Systems</source>
          , Springer, Singapore,
          <year>2021</year>
          , pp.
          <fpage>168</fpage>
          -
          <lpage>175</lpage>
          . doi:
          <volume>10</volume>
          .1007/
          <fpage>978</fpage>
          -981-16-4486-3_
          <fpage>18</fpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref4">
        <mixed-citation>
          [4]
          <string-name>
            <given-names>A.</given-names>
            <surname>Rutkowski</surname>
          </string-name>
          ,
          <string-name>
            <surname>A. Kawalec,</surname>
          </string-name>
          <article-title>Some of problems of direction finding of ground-based radars using monopulse location system installed on unmanned aerial vehicle</article-title>
          ,
          <source>Sensors</source>
          <volume>20</volume>
          (
          <year>2020</year>
          )
          <article-title>5186</article-title>
          . doi:
          <volume>10</volume>
          . 3390/s20185186.
        </mixed-citation>
      </ref>
      <ref id="ref5">
        <mixed-citation>
          [5]
          <string-name>
            <given-names>H. A.</given-names>
            <surname>Cook</surname>
          </string-name>
          ,
          <string-name>
            <surname>K. M. T. E. Kahn</surname>
            ,
            <given-names>V.</given-names>
          </string-name>
          <string-name>
            <surname>Balyan</surname>
          </string-name>
          ,
          <article-title>Radio direction-finding techniques for an unmanned aerial vehicle</article-title>
          ,
          <source>Micro-Electronics and Telecommunication Engineering, Lecture Notes in Networks and Systems</source>
          <volume>106</volume>
          (
          <year>2020</year>
          )
          <fpage>1</fpage>
          -
          <lpage>10</lpage>
          . doi:
          <volume>10</volume>
          .1007/
          <fpage>978</fpage>
          -981-15-2329-
          <issue>8</issue>
          _
          <fpage>1</fpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref6">
        <mixed-citation>
          [6]
          <string-name>
            <given-names>I.</given-names>
            <surname>Bae</surname>
          </string-name>
          ,
          <string-name>
            <surname>J. Hong,</surname>
          </string-name>
          <article-title>Survey on the developments of unmanned marine vehicles: Intelligence and cooperation</article-title>
          ,
          <source>Sensors</source>
          <volume>23</volume>
          (
          <year>2023</year>
          )
          <article-title>4643</article-title>
          . doi:
          <volume>10</volume>
          .3390/s23104643.
        </mixed-citation>
      </ref>
      <ref id="ref7">
        <mixed-citation>
          [7]
          <string-name>
            <given-names>O.</given-names>
            <surname>Ivashchuk</surname>
          </string-name>
          , et al.,
          <article-title>A configuration analysis of ukrainian flight routes network</article-title>
          ,
          <source>in: Experience of Designing and Application of CAD Systems in Microelectronics (CADSM)</source>
          ,
          <year>2021</year>
          , pp.
          <fpage>6</fpage>
          -
          <lpage>10</lpage>
          . doi:
          <volume>10</volume>
          .1109/CADSM52681.
          <year>2021</year>
          .
          <volume>9385263</volume>
          .
        </mixed-citation>
      </ref>
      <ref id="ref8">
        <mixed-citation>
          [8]
          <string-name>
            <given-names>F.</given-names>
            <surname>Yanovsky</surname>
          </string-name>
          ,
          <article-title>Inferring microstructure and turbulence properties in rain through observations and simulations of signal spectra measured with doppler-polarimetric radars, in: NATO Science for Peace and Security Series C: Environmental Security</article-title>
          , volume
          <volume>117</volume>
          ,
          <year>2011</year>
          , pp.
          <fpage>501</fpage>
          -
          <lpage>542</lpage>
          . doi:
          <volume>10</volume>
          . 1007/
          <fpage>978</fpage>
          -94-007-1636-0_
          <fpage>19</fpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref9">
        <mixed-citation>
          [9]
          <string-name>
            <given-names>Z.</given-names>
            <surname>Hussein</surname>
          </string-name>
          ,
          <string-name>
            <given-names>S.</given-names>
            <surname>Francois</surname>
          </string-name>
          ,
          <string-name>
            <given-names>B.</given-names>
            <surname>Oumaya</surname>
          </string-name>
          ,
          <string-name>
            <given-names>V.</given-names>
            <surname>Thierry</surname>
          </string-name>
          , Syloin:
          <article-title>Measuring angle of arrival of lora signals using software defined radio</article-title>
          ,
          <source>in: Proceedings of International Conference on Indoor Positioning and Indoor Navigation</source>
          , IEEE, Lloret de Mar Spain,
          <year>2021</year>
          , pp.
          <fpage>1</fpage>
          -
          <lpage>8</lpage>
          . doi:
          <volume>10</volume>
          .1109/IPIN51156.
          <year>2021</year>
          .
          <volume>9662518</volume>
          .
        </mixed-citation>
      </ref>
      <ref id="ref10">
        <mixed-citation>
          [10]
          <string-name>
            <given-names>Z.</given-names>
            <surname>Dai</surname>
          </string-name>
          ,
          <string-name>
            <given-names>Y.</given-names>
            <surname>He</surname>
          </string-name>
          ,
          <string-name>
            <given-names>V.</given-names>
            <surname>Tran</surname>
          </string-name>
          ,
          <string-name>
            <given-names>N.</given-names>
            <surname>Trigoni</surname>
          </string-name>
          ,
          <string-name>
            <given-names>A.</given-names>
            <surname>Markham</surname>
          </string-name>
          ,
          <article-title>Deepaoanet: Learning angle of arrival from software defined radios with deep neural networks</article-title>
          ,
          <source>IEEE Access 10</source>
          (
          <year>2022</year>
          )
          <fpage>3164</fpage>
          -
          <lpage>3176</lpage>
          . doi:
          <volume>10</volume>
          . 1109/ACCESS.
          <year>2021</year>
          .
          <volume>3140146</volume>
          .
        </mixed-citation>
      </ref>
      <ref id="ref11">
        <mixed-citation>
          [11]
          <string-name>
            <surname>H. B. Tuğrel</surname>
            ,
            <given-names>H.</given-names>
          </string-name>
          <string-name>
            <surname>Alakoca</surname>
            ,
            <given-names>G. K.</given-names>
          </string-name>
          <string-name>
            <surname>Kurt</surname>
            ,
            <given-names>C.</given-names>
          </string-name>
          <string-name>
            <surname>Ayyıldız</surname>
          </string-name>
          ,
          <article-title>Angle of arrival (aoa) estimation by using software defined radios</article-title>
          ,
          <source>in: Proceedings of 24th Signal Processing and Communication Application Conference</source>
          , IEEE,
          <string-name>
            <surname>Zonguldak</surname>
            <given-names>Turkey</given-names>
          </string-name>
          ,
          <year>2016</year>
          , pp.
          <fpage>1429</fpage>
          -
          <lpage>1432</lpage>
          . doi:
          <volume>10</volume>
          .1109/SIU.
          <year>2016</year>
          .
          <volume>7496018</volume>
          .
        </mixed-citation>
      </ref>
      <ref id="ref12">
        <mixed-citation>
          [12]
          <string-name>
            <given-names>Y.</given-names>
            <surname>Zheng</surname>
          </string-name>
          ,
          <string-name>
            <given-names>S.-M.</given-names>
            <surname>Tseng</surname>
          </string-name>
          ,
          <string-name>
            <surname>K.-B. Yu</surname>
          </string-name>
          ,
          <article-title>Closed-form four-channel monopulse two-target resolution</article-title>
          ,
          <source>IEEE Transactions on Aerospace and Electronic Systems</source>
          <volume>39</volume>
          (
          <year>2003</year>
          )
          <fpage>1083</fpage>
          -
          <lpage>1089</lpage>
          . doi:
          <volume>10</volume>
          .1109/TAES.
        </mixed-citation>
      </ref>
    </ref-list>
  </back>
</article>