<!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>Cybersecurity Providing in Information and Telecommunication Systems, October</journal-title>
      </journal-title-group>
    </journal-meta>
    <article-meta>
      <title-group>
        <article-title>Honeypot Security Efficiency versus Deception Solution</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <string-name>Sviatoslav Vasylyshyn</string-name>
          <xref ref-type="aff" rid="aff1">1</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Ivan Opirskyy</string-name>
          <xref ref-type="aff" rid="aff1">1</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Svitlana Shevchenko</string-name>
          <email>s.shevchenko@kubg.edu.ua</email>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <aff id="aff0">
          <label>0</label>
          <institution>Borys Grinchenko Kyiv University</institution>
          ,
          <addr-line>18/2 Bulvarno-Kudriavska str., Kyiv, 04053</addr-line>
          ,
          <country country="UA">Ukraine</country>
        </aff>
        <aff id="aff1">
          <label>1</label>
          <institution>Lviv Polytechnic National University</institution>
          ,
          <addr-line>Stepana Bandery str., 12, Lviv, 79000</addr-line>
          ,
          <country country="UA">Ukraine</country>
        </aff>
      </contrib-group>
      <pub-date>
        <year>2021</year>
      </pub-date>
      <volume>26</volume>
      <issue>2021</issue>
      <fpage>229</fpage>
      <lpage>236</lpage>
      <abstract>
        <p>Deception technology has appeared on the market of information security systems relatively recently. However, some experts still consider Security Deception to be just a more advanced “honeypot.” In this article, we will try to highlight both the similarities and fundamental differences between these two solutions. In the first part, we will tell you about honeypot, how this technology developed and what are its advantages and disadvantages. And in the second part, we will dwell on the principles of operation of platforms for creating a distributed infrastructure of false targets (DDP).</p>
      </abstract>
      <kwd-group>
        <kwd>Honeypots</kwd>
        <kwd>baits</kwd>
        <kwd>security</kwd>
        <kwd>analysis</kwd>
        <kwd>deception</kwd>
      </kwd-group>
    </article-meta>
  </front>
  <body>
    <sec id="sec-1">
      <title>1. Introduction</title>
      <p>
        Information security is a complex of technical and organizational measures and developed
documents in the broadest sense of the word. The main goal is to protect and preserve information
owned by the organization. However, information security remains an integral part of cybersecurity, a
much broader category, and includes the protection of information and data, as well as the protection
of systems, networks, etc. [
        <xref ref-type="bibr" rid="ref1 ref2 ref3">1–3</xref>
        ]. The main goals of information security also include the creation of a
set of business processes that will protect information assets regardless of how information is
formatted, whether it is in transit, processed, or is at rest, that is, stored in appropriate databases.
According to experts, the value of an object is determined primarily by what information the company
owns and how it is stored. Information security is a critical factor in ensuring the efficient conduct of
business operations, as well as in maintaining and gaining the trust of customers, both future and
existing [
        <xref ref-type="bibr" rid="ref4">4</xref>
        ]. And that is precisely why there is a problem of how to ensure the security of information
during cyberattacks. One of these ways is the use of decoys and deception solutions. Honeypot is
fundamentally different from all developments in the field of security. As a rule, all products in this
market are designed to solve a strictly defined function (it doesn't matter whether hardware or
software is involved): the firewall solves the tasks of restricting access from one network to another at
different levels, the SSH service is designed for encrypted access to operating system resources, etc.
Honeypot technology is not designed to solve a specific problem but represents a whole philosophy—
flexible, customizable in accordance with the goal [
        <xref ref-type="bibr" rid="ref5">5</xref>
        ]. The basic principle underlying honeypots is to
create traps for hackers. The very first Deception solutions were developed on the same principle.
But, modern DDPs considerably surpass honeypots, both on the functionality, and on efficiency [
        <xref ref-type="bibr" rid="ref6">6</xref>
        ].
Deception platforms include: traps (decoys, traps), lures, applications, data, databases, Active
Directory. Modern DDPs can provide ample opportunities for threat detection, attack analysis, and
automation of responses. Thus, Deception are techniques to simulate the enterprise's IT infrastructure
and mislead hackers. As a result, such platforms allow you to stop attacks before causing significant
damage to the company’s assets. Hoheypots, of course, do not have such a wide range of functionality
and such a level of automation, so their use requires more skills from employees of IS departments.
      </p>
      <p>2022 Copyright for this paper by its authors.</p>
    </sec>
    <sec id="sec-2">
      <title>2. Honeypot</title>
      <p>
        For the first time, the term “honeypots” was used in 1989 in the book “The Cuckoo’s Egg” by
Clifford Stoll, which describes the events of tracking a hacker at the Lawrence Berkeley National
Laboratory (USA). This idea was put into practice in 1999 by Lance Spitzner, an information security
specialist at Sun Microsystems, who founded the Honeynet Project. The first honeypots were very
resource intensive, difficult to set up and maintain [
        <xref ref-type="bibr" rid="ref10 ref7 ref8 ref9">7–10</xref>
        ].
      </p>
      <p>Let’s take a closer look at what honeypots and honeynets are. Honeypots are separate hosts, the
purpose of which is to attract intruders to penetrate the company's network and try to steal valuable
data, as well as expand the network coverage. A honeypot (literally translated as “a keg of honey”) is
a special server with a set of different network services and protocols such as HTTP, FTP, etc.
(Fig. 1).</p>
      <p>
        If we combine several honeypots into a network, then we get a more efficient honeynet system,
which is an emulation of the company's corporate network (web server, file server, and other network
components). Such a solution allows us to understand the strategy of actions of the attackers and
mislead them. A typical honeynet typically runs in parallel with, and completely independent of, a
production network. Such a "network" can be published on the Internet via a separate channel, and a
separate range of IP addresses can also be allocated for it [
        <xref ref-type="bibr" rid="ref11">11</xref>
        ].
      </p>
      <p>The obvious benefit of honeynets is that they mislead intruders, wasting their energy, resources
and time. As a result, instead of real targets, they attack false ones and can stop attacking the network
without achieving anything. Most often, honeynets technologies are used in government agencies and
large corporations, financial organizations, since these structures are the targets for large cyber
attacks.</p>
      <p>Why are honeypots and honeynets not the best solutions to counter attacks today? It should be
noted that attacks are becoming more large-scale, technically complex and capable of causing serious
damage to the IT infrastructure of an organization, while cybercrime has reached a completely
different level and represents highly organized shadow business structures equipped with all the
necessary resources. To this must be added the “human factor” (errors in the settings of software and
hardware, actions of insiders, etc.), so the use of only technologies to prevent attacks is no longer
enough at the moment.</p>
      <p>Below are the main limitations and disadvantages of honeypots (honeynets):
 Honeypots were originally designed to identify threats that are outside the corporate network,
are intended rather to analyze the behavior of intruders and are not designed to quickly
respond to threats.
 Attackers have usually learned how to recognize emulated systems and avoid honeypots.
 Honeynets (honeypots) have an extremely low level of interactivity and interaction with other
security systems, as a result of which, using honeypots, it is difficult to obtain detailed
information about attacks and attackers, and therefore, to effectively and quickly respond to
information security incidents. Moreover, information security specialists receive a large
number of false alerts about threats.
 In some cases, hackers can use a compromised honeypot as a starting point to continue their
attack on an organization's network.
 Problems often arise with the scalability of honeypots, high operational load and
configuration of such systems (they require highly qualified specialists, do not have a
convenient management interface, etc.). There are great difficulties in deploying honeypots in
specialized environments such as IoT, POS, cloud systems, etc.</p>
    </sec>
    <sec id="sec-3">
      <title>3. Deception Technology</title>
      <p>
        Having studied all the advantages and disadvantages of honeypots, we come to the conclusion that
a completely new approach to responding to information security incidents is needed in order to
develop a quick and adequate response to the actions of attackers. And such a solution is Cyber
deception (Security deception) technologies [
        <xref ref-type="bibr" rid="ref12">12</xref>
        ].
      </p>
      <p>The terminology “Cyber deception,” “Security deception,” “Deception technology,” “Distributed
Deception Platform” (DDP) is relatively new and appeared not so long ago. In fact, all of these terms
mean the use of “deception technologies” or “techniques to imitate IT infrastructure and misinform
attackers.” The simplest Deception solutions are the development of honeypots ideas, only at a more
technologically advanced level, which involves more automation in detecting and responding to
threats. However, there are already serious DDP-class solutions on the market, which imply ease of
deployment and scalability, and also have a serious arsenal of “traps” and “baits” for attackers.</p>
      <p>How does the “Distributed Deception Platform” work? After the deployment of DDP, the IT
infrastructure of the organization will be built as if from two layers: the first layer is the real
infrastructure of the company, and the second is an "emulated" environment consisting of decoys,
traps and decoys. lures) that are located on real physical devices on the network (Fig. 2).</p>
      <p>For example, an attacker can detect false databases with “confidential documents,” fake credentials
of supposedly “privileged users”—all these are false targets, they can interest the attackers, thereby
diverting their attention from the true information assets of the company.</p>
      <p>DDP is a novelty in the market of information security products, these solutions are only a few
years old and so far only the corporate sector can afford them. But small and medium businesses will
soon be able to take advantage of Deception, leasing DDPs from specialized providers “as a service.”
This option is even more convenient, since there is no need for your own highly qualified personnel.</p>
      <p>The main benefits of Deception technology are shown below:
 Authenticity (authenticity). Deception technology is able to reproduce a completely authentic
IT environment of a company, qualitatively emulating operating systems, IoT, POS, specialized
systems (medical, industrial, etc.), services, applications, credentials, etc. Decoys are carefully
mixed with the production environment, and an attacker would not be able to identify them as
honeypots.
 Implementation. DDPs use machine learning (ML) in their work. ML provides simplicity,
customization, and efficiency in implementing Deception. “Traps” and “decoys” are very
quickly updated, involving the attacker in the “false” IT infrastructure of the company, and in
the meantime, advanced analysis systems based on artificial intelligence can detect active
hacker actions and prevent them (for example, an attempt to access Active Directory based on
fraudulent accounts).
 Ease of operation. Modern “Distributed Deception Platforms” are easy to maintain and manage.</p>
      <p>Typically, they are managed through a local or cloud console, there is the possibility of
integration with the corporate SOC (Security Operations Center) through the API and with
many existing security controls. DDP service and operation does not require the services of
highly qualified information security experts.
 Scalability. Security deception can be deployed in physical, virtual and cloud environments.</p>
      <p>DDPs successfully work with specialized environments such as IoT, ICS, POS, SWIFT, etc.
Enhanced Deception platforms can project “trickery” technologies into remote offices, isolated
environments, without the need for additional full platform deployment.
 Interaction. Using effective and attractive decoys that are based on real operating systems and
cleverly placed among the real IT infrastructure, the Deception platform collects a wealth of
information about the attacker. DDP then delivers threat alerts, generates reports, and
automatically responds to information security incidents.
 The starting point of the attack. In modern Deceptions, traps and baits are placed within the
range of the network, rather than outside of it (as is the case with honeypots). This model of
deploying traps prevents an attacker from using them as a launching point for an attack on a
company's real IT infrastructure. In more advanced solutions of the Deception class, there are
traffic routing capabilities, thus it is possible to direct all traffic of the attackers through a
dedicated connection. This will allow you to analyze the activity of intruders without risking
the company's valuable assets.
 The persuasiveness of "technology of deception". At the initial stage of an attack,
cybercriminals collect and analyze data about the IT infrastructure, and then use it to move
horizontally across the corporate network. With the help of "technologies of deception" the
attacker will surely fall into "traps" that will lead him away from the real assets of the
organization. DDP will analyze potential paths to access credentials on the corporate network
and provide the attacker with "decoys" instead of the real credentials. These capabilities have
been sorely lacking in honeypot technology.</p>
    </sec>
    <sec id="sec-4">
      <title>4. Honeypot and Deception Comparison</title>
      <p>
        We come to the most interesting point of our research. Let’s try to highlight the main differences
between Deception and Honeypot technologies. Despite some similarities, these two technologies are
still very different, ranging from the fundamental idea to the efficiency of work [
        <xref ref-type="bibr" rid="ref13 ref14">13,14</xref>
        ].
 Different basic ideas. As we wrote above, honeypots are installed as “decoys” around valuable
company assets (outside the corporate network), thus trying to distract intruders. While
honeypots are based on an understanding of an organization’s infrastructure, honeypots can be
a pivotal point to launch an attack on a company’s network. Deception technology is designed
with an attacker's point of view in mind and allows you to identify an attack at an early stage,
thus, information security specialists get a significant advantage over attackers and gain time.
 Attraction VS Obfuscation. When using honeypots, success depends on attracting the attention
of attackers and further motivating them to navigate to the target in the honeypot. This means
that the attacker still has to get to the honeypot before you can stop him. Thus, the presence of
attackers on the network can last for several months or more, and this will lead to data leakage
and damage. DDPs qualitatively imitate the real IT infrastructure of a company, the purpose of
their implementation is not just to attract the attention of an attacker, but to confuse him so that
he wasted time and resources, but did not gain access to the real assets of the company.
 “Limited scalability” VS “Automatic scalability.” As noted earlier, honeypots and honeynets
have scaling issues. It is difficult and expensive, and in order to increase the number of
honeypots in a corporate system, you will have to add new computers, operating systems,
purchase licenses, and allocate IP. Moreover, it is also necessary to have qualified personnel to
manage such systems. Deception platforms are automatically deployed as the infrastructure
scales, without significant overhead.
 “A large number of false positives” VS “no false positives.” The essence of the problem is that
even a simple user may encounter a honeypot, therefore the “flip side” of this technology is a
large number of false positives, which distracts information security specialists from their
work. "Decoys" and "traps" in DDP are carefully hidden from the common user and are
designed only for the attacker, so each signal from such a system is a notification of a real
threat, and not a false alarm.
      </p>
      <p>As an example of Honeypot and Deception technology comparison, please check Table 1. Please
note that this is rough comparison chart made by looking over the standard possibilities provided by
each platform and don’t includes unique solutions provided by private companies. Also this chart uses
format from the previous tables that can be seen. It is done for the sake of formatting and for the
purpose to create a standard for this kind of comparison between similar but different technologies.
This way we will have a basic chart using all actual and use-case technologies in one place.
Deception technology</p>
      <p>Linux/ Windows
Intelligence: Active
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
AWS/Azure
+
+
+
+
+
+
+
+
+
+</p>
      <p>For a scientific point of view we conducted a simple test by facing both out-of-the-box systems
before written detection bot. As far as we can see form the Fig. 3, the Deception system has lower
percentage in comparison to Honeypot to be detected. This tendentious fracture is secured on every
try of the test. On average Honeypot is by 7% more detectable. The comparison you can see from the
Fig. 4.</p>
    </sec>
    <sec id="sec-5">
      <title>5. Conclusions</title>
      <p>Deception technology is a huge improvement over the older Honeypots technology. In essence,
DDP has become a comprehensive security platform that is easy to deploy and manage.</p>
      <p>Modern platforms of this class play an important role in accurate detection and effective response
to network threats, and their integration with other components of the security stack increases the
level of automation, increases the efficiency and effectiveness of incident response. Deception
platforms are based on authenticity, scalability, ease of management and integration with other
systems. All this gives a significant advantage in the speed of response to information security
incidents.</p>
      <p>Also, based on observations of pentests of companies where the system was implemented, it can
be concluded that even experienced pentesters often cannot recognize the baits in the corporate
network and are defeated, falling for the set traps. This fact once again confirms the effectiveness of
Deception and the great prospects that open up for this technology in the future.
6. References</p>
    </sec>
  </body>
  <back>
    <ref-list>
      <ref id="ref1">
        <mixed-citation>
          [1]
          <string-name>
            <given-names>V.</given-names>
            <surname>Lakhno</surname>
          </string-name>
          , et al.,
          <article-title>Management of information protection based on the integrated implementation of decision support systems, Eastern-european journal of enterprise technologies</article-title>
          .
          <source>Information and controlling system</source>
          , vol.
          <volume>5</volume>
          , no.
          <volume>9</volume>
          (
          <issue>89</issue>
          ),
          <fpage>36</fpage>
          -
          <lpage>41</lpage>
          ,
          <year>2017</year>
          . https://doi.org/ 10.15587/
          <fpage>1729</fpage>
          -
          <lpage>4061</lpage>
          .
          <year>2017</year>
          .
          <volume>111081</volume>
          .
        </mixed-citation>
      </ref>
      <ref id="ref2">
        <mixed-citation>
          [2]
          <string-name>
            <given-names>V.</given-names>
            <surname>Dudykevych</surname>
          </string-name>
          , et al.,
          <article-title>A multicriterial analysis of the efficiency of conservative information security systems, Eastern-european journal of enterprise technologies</article-title>
          .
          <source>Information and controlling system</source>
          , vol
          <volume>3</volume>
          , no 9
          <issue>(99</issue>
          ),
          <fpage>6</fpage>
          -
          <lpage>13</lpage>
          ,
          <year>2019</year>
          . https://doi.org/10.15587/
          <fpage>1729</fpage>
          -
          <lpage>4061</lpage>
          .
          <year>2019</year>
          .166349
        </mixed-citation>
      </ref>
      <ref id="ref3">
        <mixed-citation>
          [3]
          <string-name>
            <surname>M.-D. McLaughlin</surname>
            ,
            <given-names>G.</given-names>
          </string-name>
          <article-title>Janis, Challenges and best practices in information security management</article-title>
          ,
          <source>MIS Quarterly Executive 17.3</source>
          (
          <year>2018</year>
          ):
          <fpage>237</fpage>
          -
          <lpage>262</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref4">
        <mixed-citation>
          [4]
          <string-name>
            <given-names>V.</given-names>
            <surname>Susukailo</surname>
          </string-name>
          ,
          <article-title>Cybercrimes investigation via honeypots in cloud environments</article-title>
          ,
          <source>CEUR Workshop Proceedings</source>
          , pp.
          <fpage>91</fpage>
          ,
          <year>2021</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref5">
        <mixed-citation>
          [5]
          <string-name>
            <given-names>R. C.</given-names>
            <surname>Joshi</surname>
          </string-name>
          ,
          <string-name>
            <given-names>A.</given-names>
            <surname>Sardana</surname>
          </string-name>
          ,
          <article-title>Honeypots: A new paradigm to information security</article-title>
          , in Honeypots: A New Paradigm to Information Security,
          <fpage>1</fpage>
          -
          <lpage>323</lpage>
          ,
          <year>2011</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref6">
        <mixed-citation>
          [6]
          <string-name>
            <given-names>D.</given-names>
            <surname>Zhuravchak</surname>
          </string-name>
          ,
          <article-title>Сreation of a system for preventing the spread of extractor viruses using the python programming language and auditd utility based on operational system. Electronic professional scientific publication “Cybersecurity: Education, Science</article-title>
          , Technology,”
          <volume>4</volume>
          (
          <issue>12</issue>
          ),
          <fpage>108</fpage>
          -
          <lpage>116</lpage>
          ,
          <year>2021</year>
          . https://doi.org/10.28925/
          <fpage>2663</fpage>
          -
          <lpage>4023</lpage>
          .
          <year>2021</year>
          .
          <volume>12</volume>
          .108116
        </mixed-citation>
      </ref>
      <ref id="ref7">
        <mixed-citation>
          [7]
          <string-name>
            <given-names>S.</given-names>
            <surname>Udhani</surname>
          </string-name>
          ,
          <string-name>
            <given-names>A.</given-names>
            <surname>Withers</surname>
          </string-name>
          ,
          <string-name>
            <given-names>M.</given-names>
            <surname>Bashir</surname>
          </string-name>
          ,
          <article-title>Human vs bots: Detecting human attacks in a honeypot environment</article-title>
          . in A Varol,
          <string-name>
            <given-names>M</given-names>
            <surname>Karabatak</surname>
          </string-name>
          ,
          <string-name>
            <given-names>C</given-names>
            <surname>Varol</surname>
          </string-name>
          &amp; S Teke (eds),
          <source>7th International Symposium on Digital Forensics and Security</source>
          ,
          <string-name>
            <surname>ISDFS</surname>
          </string-name>
          <year>2019</year>
          .,
          <volume>8757534</volume>
          , 7th
          <source>International Symposium on Digital Forensics and Security</source>
          ,
          <string-name>
            <surname>ISDFS</surname>
          </string-name>
          <year>2019</year>
          ,
          <article-title>Institute of Electrical and Electronics Engineers Inc</article-title>
          .,
          <source>7th International Symposium on Digital Forensics and Security</source>
          ,
          <string-name>
            <surname>ISDFS</surname>
          </string-name>
          <year>2019</year>
          , Barcelos, Portugal,
          <volume>6</volume>
          /10/19. https://doi.org/10.1109/ISDFS.
          <year>2019</year>
          .8757534
        </mixed-citation>
      </ref>
      <ref id="ref8">
        <mixed-citation>
          [8]
          <string-name>
            <given-names>C.</given-names>
            <surname>Moore</surname>
          </string-name>
          ,
          <article-title>Detecting ransomware with honeypot techniques</article-title>
          ,
          <source>Proceedings 2016 Cybersecurity and Cyberforensics Conference</source>
          ,
          <string-name>
            <surname>CCC</surname>
          </string-name>
          <year>2016</year>
          ,
          <volume>77</volume>
          .
        </mixed-citation>
      </ref>
      <ref id="ref9">
        <mixed-citation>
          [9]
          <string-name>
            <given-names>S.</given-names>
            <surname>Litchfield</surname>
          </string-name>
          , et al.,
          <article-title>Rethinking the Honeypot for Cyber-Physical Systems</article-title>
          ,
          <source>IEEE Internet Computing</source>
          , vol.
          <volume>20</volume>
          , no.
          <issue>5</issue>
          ,
          <fpage>9</fpage>
          -
          <lpage>17</lpage>
          ,
          <year>2016</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref10">
        <mixed-citation>
          [10]
          <string-name>
            <given-names>Z. A.</given-names>
            <surname>Khan</surname>
          </string-name>
          , U. Abbasi,
          <article-title>Reputation Management Using Honeypots for Intrusion Detection in the Internet of Things</article-title>
          , Electronics,
          <volume>9</volume>
          ,
          <year>2020</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref11">
        <mixed-citation>
          [11]
          <string-name>
            <given-names>M.</given-names>
            <surname>Akiyama</surname>
          </string-name>
          , et al.,
          <article-title>HoneyCirculator: distributing credential honeytoken for introspection of web-based attack cycle</article-title>
          .
          <source>International Journal of Information Security</source>
          ,
          <year>2017</year>
          . https://doi.org/10.1007/s10207-017-0361-5
        </mixed-citation>
      </ref>
      <ref id="ref12">
        <mixed-citation>
          [12]
          <string-name>
            <given-names>V.</given-names>
            <surname>Gandotra</surname>
          </string-name>
          ,
          <string-name>
            <given-names>A.</given-names>
            <surname>Singhal</surname>
          </string-name>
          ,
          <string-name>
            <given-names>P.</given-names>
            <surname>Bedi</surname>
          </string-name>
          ,
          <article-title>Threat-Oriented Security Framework: A Proactive Approach in Threat Management</article-title>
          .
          <source>Procedia Technology</source>
          ,
          <volume>4</volume>
          ,
          <fpage>487</fpage>
          -
          <lpage>494</lpage>
          ,
          <year>2012</year>
          . https://doi.org/10.1016/ j.protcy.
          <year>2012</year>
          .
          <volume>05</volume>
          .078
        </mixed-citation>
      </ref>
      <ref id="ref13">
        <mixed-citation>
          [13]
          <string-name>
            <given-names>V.</given-names>
            <surname>Buriachok</surname>
          </string-name>
          ,
          <string-name>
            <given-names>V.</given-names>
            <surname>Sokolov</surname>
          </string-name>
          ,
          <string-name>
            <given-names>P.</given-names>
            <surname>Skladannyi</surname>
          </string-name>
          ,
          <article-title>Security rating metrics for distributed wireless systems</article-title>
          ,
          <source>in: Workshop of the 8th International Conference on “Mathematics. Information Technologies. Education,” Modern Machine Learning Technologies and Data Science (MoMLeT and DS)</source>
          , vol.
          <volume>2386</volume>
          ,
          <fpage>222</fpage>
          -
          <lpage>233</lpage>
          ,
          <year>2019</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref14">
        <mixed-citation>
          [14]
          <string-name>
            <given-names>J.</given-names>
            <surname>Onaolapo</surname>
          </string-name>
          , E. Mariconti, G. Stringhini,
          <article-title>What Happens After You Are Pwnd: Understanding The Use Of Leaked Account Credentials In The Wild</article-title>
          .
          <source>Proceedings of the 2016 ACM on Internet Measurement Conference, IMC 16</source>
          ,
          <year>2016</year>
          . https://doi.org/10.1145/2987443.2987475
        </mixed-citation>
      </ref>
    </ref-list>
  </back>
</article>