MIME-Version: 1.0 Content-Type: multipart/related; boundary="----=_NextPart_01D63B8F.86552F20" Este documento es una página web de un solo archivo, también conocido como "archivo de almacenamiento web". Si está viendo este mensaje, su explorador o editor no admite archivos de almacenamiento web. Descargue un explorador que admita este tipo de archivos. ------=_NextPart_01D63B8F.86552F20 Content-Location: file:///C:/2554122F/06electronic-methods-for-obtaining-ip_GabrilMoreano.htm Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset="us-ascii"
Electronic methods for obtaining IP addresses thro=
ugh
webmail and social networks
M&eacut= e;todos electrónicos para obtención de direcciones IP a través= de webmail y redes sociales
Gabriel Vinicio Moreano Sánchez. [1= ], Edgar Vinicio Ávalos Yuque . [2= ], Victor Hugo Benitez= Bravo. [3= ] & Alvaro Gabriel Benitez Bravo. [4= ]
Reci= bido: 12-02-2020 / Revisado: 05-03-2020 /Aceptado: 12-04-2020/ Publicado: 05-06-2020<= /p>
Abstrac=
t.
&=
nbsp; &nbs=
p; =
DOI: https://doi.org/10.33262/concienciadigita=
l.v3i2.2.1247
=
Obtaining data is a fundamental stage in a computer
security process both to identify violators or to detect vulnerabilities in=
a
certain system. For tasks such as investigation or intelligence, among the =
most
important information that can be collected from an entity are the IP addre=
sses
of important people, suspects or investigation targets. Obtaining this
information for legal and ethical purposes is currently more complex because
service providers mask this data in order to protect the identity of their
users. In order to obtain the IP address, it is intended to apply different
electronic and automated techniques that serve to modify documents and cover
them up so that, when opened by the recipient, they reveal their IP address,
which will be automatically registered in the logs of the web server that is
implemented in this research. These approaches will be considered: sending =
an
invitation HTTP link directly and sending an Office document with invisible
characters. The scope of the work described is important, since it covers
proposed techniques for detecting an IP address using various methods
evaluating the efficiency and effectiveness of each test. This will allow t=
he
forensic analyst or investigator to obtain a person's possible location by
consulting with internet providers.
Resumen.
La obtención de datos es una etapa fundamen=
tal
en un proceso de seguridad informática tanto para identificar
infractores o para detectar vulnerabilidades en un determinado sistema. Para
tareas como investigación o inteligencia, entre la infor-mación
más importante que se puede levantar de un ente están las
direcciones IP de personas importantes, sospechosas u objetivos de
investigación. Obtener esta información con fines legales y
éticos, actualmente es más complejo debido a que los proveedo=
res
de servicios enmascaran este dato a fin de proteger la identidad de sus
usuarios. Para obtener la dirección IP se pretende aplicar diferentes
técnicas electrónicas y automatizadas que sirvan para modific=
ar
documentos y encubrirlos de modo que, al ser abiertos por el destinatario, =
nos
revelen su dirección IP, la cual quedará registrada
automáticamente en los logs del servidor web que se implementa en es=
ta
investigación. Se tomará en cuenta estos enfoques: el enviar
directamente un enlace HTTP de invitación y enviar un documento de
Office con caracteres invisibles. El ámbito de aplicación del
trabajo descrito es importante, pues abarca técnicas propuestas para=
la
detección de una dirección IP mediante varios métodos
evaluando la eficiencia y efectividad de cada prueba. Esto permitirá=
al
analista forense o investigador, obtener la ubicación posible de una
persona al consultarla con los proveedores de internet.
Palabras claves:=
DNS, Dirección IP, W=
ebmail,
Redes sociales, Analista Forense.
<=
span
lang=3DEN-US style=3D'font-size:12.0pt;line-height:115%;font-family:"Times =
New Roman",serif;
mso-ansi-language:EN-US;mso-fareast-language:EN-US;mso-bidi-language:SI-LK'=
>Introducción.
In modern times the advancement of technology is
exponential. Everything changes minute by minute, and this has generated gr=
eat
advances and changes in society such as more agile markets, online educatio=
n at
all types of levels and borders, more continuous communication with our lov=
ed
ones and others, but it has also caused new problems such as: crime, altera=
tion
of public and private information, falsification of documents, large-scale
scams, disappearance of bank accounts, etc. In this work we focus on the
merciless attacks on cyber media that cause a global scandal by revealing
private or confidential information, a situation that no one is except for.=
We
are witnesses to the birth of viruses, malwares that attack the computer
systems of the world causing irreparable damage to storage infrastructure, =
file
integrity in a company or other computer users.
This insecurity has resulted in new social ills and the enlargement of others that already exist, such as: cyberbullying, child pornography, child trafficking, drug trafficking, etc. In this work the way= s of detecting the person or persons located on the other side of the computer or mobile device are considered; Identify them so that justice can proceed aga= inst them. In Latin America, several surveys indicate that social networks are t= he main route of harassment, that is, traditional harassment has migrated towa= rds information and communication technologies and especially social networks.<= o:p>
According to the company Eset=
- Latin America, an informatics security company, in 2013, cyberbullying sh=
owed
the following statistics: 30.7% of adolescents in the region aged 12 - 20 w=
ere
victims of Internet harassment. The main route of these attacks was social
network like Facebook, Twitter and Google Plus with 82.7%.
Figure
1. Ways of
harassment
Prepared by:= Research Group.<= o:p>
In (De Juventud & De, 2017) the impact of bull=
ying
and cyberbullying is widely shown, in a doctoral thesis (Gonzales
García, 2015) the author shows us an approach to antisocial behavior=
in
cyberspace and how this generates a high risk over the most innocent people=
and
also shows us certain strategies to counteract this problem. In (Garc&iacut=
e;a
Maldonado, Joffre - Velásquez, Jes&uacut=
e;s
Martínez - Salazar, & Llanes - Casti=
llo,
2011, Ximhai, 2014) we find a specific summary =
on how
this problem affects the most exposed people who are adolescents and childr=
en,
finally (Herrera - López, Romera, Ortega=
-
Ruiz, & Ortega - Ruiz, 1996) give us a vision about the impact of this =
type
of unethical harassment in our region.
In Ecuador since 2014, date on which the Comprehen=
sive
Organic Criminal Code (COIP) came into force, cybercrime is considered and
sanctioned, and which is committed through the use of computers, communicat=
ion
computerized devices or an informatics system.
According to the National Police of Ecuador "=
in
the country, 85% of computer crimes occur due to carelessness of users when
accessing social networks, using smartphones, email or using passwords, thi=
s is
gives false advertising posts or malicious emails requesting password chang=
es
and others. Also, excessively exposing information on social networks such =
as
Facebook, Twitter, Instagram and others facilitate the work of scammers,
extortionists, kidnappers, and those who are involved with child pornograph=
y,
"the excess information on the networks causes much People have access=
to
that information and that may violate emails, bank accounts and others by
skipping security questions with dates, names, studies or jobs of the victi=
m.
From January to May 2016, 530 complaints for computer crimes were presented=
to
the Ecuadorian State Attorney General's Office, the majority corresponding =
to
fraudulent appropriation by electronic means. (Arroyo J. Richard, 2016). The
computer crimes that occur most frequently in Ecuador are electronic fraud =
and
child pornography; For this reason, this research is proposed in order to
mitigate this problem.
At an industrial level, it is also important=
to
show a high level of safety in order to protect valuable data and informati=
on,
in (Oscar Fernando Castellanos, Montañez,
& Fonseca, 2007) a summary on how digital marketing and a virtual model
strengthens the industry was identified. In (Sarmiento, Guerrero, &
PYME’s are also the target of these
digital criminals and specifically attack the servers of these entities in =
an
effort to violate their banking systems and obtain information from their
clients to assign them as new victims, in (Johanna Martínez Molina et
al., 2009) the authors propose a double firewall filtering technique to avo=
id
these attacks on servers, a system that will help stop certain attacks but =
will
not release information from those responsible.
To improve security systems in all areas of
society, residential, commercial and industrial areas, processes for verify=
ing
security levels have been developed, that is, there are companies that are
dedicated to commercializing services in computer breaches in order to to find errors in the systems and correct them, and t=
here
are other types of companies that are in charge of certifying security leve=
ls,
in (Bracho et al., 2017) a form of computer aud=
it is
presented according to the OSSTMMv3 methodology, a very respectable approac=
h to
verify the security of certain computer systems, while (Ibarra & Electrónico, 2004) presents a general approach=
to
how technology auditing helps public and private companies and legal and
natural persons to maintain an adequate level of digital security, from
Likewise, conductive rules for the use of computer tools are shown in order=
to
keep our information safe.
This work focuses on obtaining information f=
rom
cyber attackers not necessarily after they have carried out a wrongdoing, b=
ut
rather obtaining information from people who are presumably digital crimina=
ls,
specifically seeking to obtain their physical location by identifying the IP
address from where If fraudulent operations are taking place, the automated
identification process attacks the target with various automated digital to=
ols
in order to mislead the target and succeed in the task.
Methodology
In order to carry out the experiment, we worked on=
the
installation and configuration of a web server to store the logs of people,
where through social engineering and by clicking on a link, when opening an
office document, those become the object of this investigation.
Web
Server Configuration
For the present investigation the Microsoft Azure
cloud platform is manipulated, along with the operating system as a virtual
Windows Server 2016, with internet information Server (IIS). The configurat=
ion
of the web server is carried out under the following considerations: basic
configuration name WS2016, HDD virtual machine disk type, user name and
password. To install the server we considered a near place to Ecuador and t=
he
other fields by default. The size of the virtual machine is allocated two
cores, RAM of 7 GB, 14 GB of disk storage. The most important step in this
configuration is: the network settings where NAT is configured to port 80, =
to
make the web server accessible from all over the world.
Once the virtual machine was created, you can see =
its
structure, with buttons to connect, start, restart, stop among other options
similar to those found at the desktop virtualization tools. A remote connec=
tion
is used to connect to the virtual operating system. The connection to Micro=
soft
Server 2016 Datacenter is through a remote desktop. When the configuration =
is
done, a file is downloaded, which is double-clicked to enter the credential=
s.
The web server is configured with the IIS installation manager, with the
minimum requirements being used in this investigation. When entering the pu=
blic
IP assigned by the Microsoft Azure platform in a browser of the web server =
or
any browser available in any location: the initial page of the web server is
loaded.
Web Bug
In informatics, a Web Bug is considered as a web b=
eacon
or a tracker, an invisible image to the human eye with a size no larger tha=
n /
x / pixels; but that can be inserted on a web page or in an e-mail message.=
It
is used in order to control whether a person has read an e-mail or visited a
web page (ideal for the marketing of a company). This allows a control of v=
iews
of a specific page, as well as various functions such as web analysis. The =
Web
Bug designed for this research will recover the public IP of visitors to the
web page configured for this study.
Web Page
A new interface is customized creating a web page = with the name "TFM MSI". Therefore, we proceed with the verification by entering the public IP address 192.232.34.208; It also adds the Web Bug that will serve as a tracker. In the same way, in the testing stage, the address= of the image is sent by social networks or by email using social engineering.<= o:p>
Web
server Logs
Under the observatio=
n of
what IP addresses were opened the image or Web bug, with url
http: //191.232.34.208/images/viru=
s.jpg,
stored in the web server; it is necessary to watch the folder logs C:/inetpub/logs/logfiles/W3SVC2, =
where
it is shown the year, month, day, exact time, archive name, used port, and =
most
importantly, The public IP address of the device which is important to stre=
ss
on the folder´s name /virus.=
jpg
just to set an example of this research. This points out precisely that such
folder may hold malicious programs in a real environment the archives names
that have been used should not raise any suspicions.
In order to acquire =
the
reports in real time, the power shell tool is used from Windows server 2016
Datacenter; under the following command c:/inetpub/logs/LogFiles/W3SVC2>
Get-content./filename.log-Wait.
Word 2016 Document
Configuration
For the next experimentation, the Word 2016 versio=
n is
used, with the option of quick elements. Choosing and inserting a field nam=
ed
“Include picture” =
referencing
an image stored on the web server. It is important to emphasize that, by me=
ans
of this technique, the complete image is not inserted in the document, but =
only
a link to it. In such a way that the document to be opened, the system will
connect to the web server, in order to load the image and be able to show i=
t.
In this way, we make sure to get a trace of the object each time you open t=
he
document.
IncludePicture =
Field
The referenced address is http://191.232.34.208/images/virus.jpg. Note that this test wor= ks with any version of Word installed on any operating system. This test is do= ne by attaching the file to an email to send it later to the target, person or target, to perform the operation we enter text in the document, for later s= end it to the target. As you can see, the Web Bug appears before the text. Alth= ough in this case, we have used a black dot in order to make it visible in the tests. In a real case, a transparent or white pixel would be used, so that = it would not be visible to the user, but it would still have the same effect.<= o:p>
Results
Graphic
1. localizer IP
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Prepared by: Research Group.<= o:p>
Test son Social Network
Facebook
Sending the URL to the target. It is used for this
test in a web page stored in Windows Server 20016 Datacenter on the Microso=
ft
Azure cloud. The Web Bug is configured in size 1x1, the URL contains the Web
Bug to be sent to the target. For these tests, shipments are made between t=
wo
accounts: the one that will send the messages created expressly for this
purpose, and the one that will receive the messages, which is the account of
the objective of the investigation. The Web Bug is included as a tracker http://191.232.34.208/images/PremioAu=
o.jpg.
A fake Facebook account with random data is then configured. As next step by
remote desktop, you enter the server to modify the file index.html with the web Bug.
Fig=
ure 2.
Web Bug draw
Prepared by:= Research Group.<= o:p>
Test 1. Facebook share a
publication
To check this method, a publication is created with
the link http://191.232.34.208/ima=
ges/PremioAuo.jpg,
which shows the draw of a car that will be made soon, the objective of this
test is to prove the scenario, check if the method proposed is valid and
analyze the results obtained by the people who entered the website that
contains the Web Bug. Prior to the publication of this publication in the
created account, it is added to as many people as possible, obtaining in a =
few
hours a large number of accepted friend requests. The Logs shown in Figure 3
indicate the entry to the page with the tracker, the high interaction with =
only
seconds of difference stands out in this test.
Fig=
ure 3.
Logs test auto raffle
Prepared by:= Research Group.<= o:p>
In the same way, it is appreciated that the income=
is
made from several IP addresses, these addresses correspond to Facebook data
centers and do not correspond to those of the users as shown in Figure 4 wh=
en
reviewing the WHOIS records with the address 173.252.124.21. therefore, we can conclude that this is not a v=
alid
method to obtain the IP address of our objective since the connection to our
web server is carried out by the Facebook systems themselves, which in turn
will later serve this content to the recipient. In short, by this method, w=
e do
not obtain an IP address that identifies the target.
Figure
4. WHOIS Records=
Prepared by: Research Group.<= o:p>
Test
2. Facebook send URL private message
For the next test the individual sending of the UR=
L is
done by direct message. Obviously, in a real case it would be necessary to
carry out a study of the objective in order to elaborate a publication adju=
sted
to its interests, in order to obtain greater efficiency.
Figure
5. Web server lo=
gs
direct message
Prepared by:=
Research Group.1=
The target is identified, and the URL is sent by
private message. It is appreciated that the target has seen the message at =
12:04, to determine the IP addres=
s, it
is necessary to go to the Server Logs (fig 5), thus verifying that the time=
is
the same as registered by the social network when opening the attached link=
and
IP address 69.63.188.215.
Fig=
ure 6.
Localización IP men=
saje
directo
Prepared by: Research Group.<= o:p>
Continuing with the development of the test, the d=
iscovered
address is entered in a geo locator of IP addresses showing the information=
of
Figure 6.
The data obtained do not correspond to the
possible location of the objective based on the study carried out, as the
Internet service provider or ISP indicates that it is Facebook, in these ca=
ses
the location is questionable, as shown in figure 7 when geo is localized the
same IP address 69.63.188.215.=
Figure
7. Wrong locatio=
n IP
address
Prepared by:= Research Group.<= o:p>
With this experimentation it is concluded that the=
IP
addresses found correspond to the Facebook servers which download the
photograph and these are registered in the Logs, once again these IP addres=
ses
do not correspond directly to the users who opened the link, and therefore =
he
deduces that this method is not valid.
Tests with Webmails
In this section tests are carried out with <=
span
class=3DSpellE>Webmails sending emails between Gmail and Outlook acc=
ounts,
for this purpose it uses URLs that reference images or Web Bugs, in additio=
n to
a file to capture the Logs and later georeference IP
addresses.
Test 1. URL directly linki=
ng
The first test is to reference the URL with the Web
Bug http://191.232.34.208/images/P=
remioAuto.jpg
in an email message in Gmail and send it to another Outlook email account. =
Use
the option to upload photos then add the URL, by pressing the button insert=
the
image will be referenced, to better convince the target or person who recei=
ves
the mail in the subject field can be placed a legend, after this the sendin=
g is
made. When opening the email in Outlook, it is observed that the message has
arrived at 18:39.
In the logs of the web server, the address 191.232.34.208 can be seen, that is, the place where the email was opened.<= o:p>
Figure
8. Test 1 Log wh=
en
opening mail in Outlook
Prepared by:= Research Group.<= o:p>
By georeferencing the IP shows the location of the=
web
server that opened (Figure 9) the mail however the information is not
conclusive, they are Microsoft machines that download the image, and serve =
the
recipient. There is no contact between the target and the server, so this t=
est
is not valid.
Fig=
ure 9.
Mail opening location
=
Prepared
by: Research Group.
Test 2. Attaching Word
Document
This test consists of sending a name attachment test 2tfm.doc between email accou=
nts
Gmail, the previously configured document contains a WEB Bug in the address=
http://191.232.34.208/images/virus.jpg
referencing the Web server to register logs when opened. It also includes t=
he
URL http://191.232.34.208/images/PremioAuto.jpg
in the email message that just like the previous link contains a Web Bug wh=
ich
will allow you to determine what time the mail was opened in the registered
Logs.
The sent is done from the account vinycioavalos@gmail.com to the post office 3hchocoatl@gmail.com. mail opens 3hcho-coatl@gmail.com and the message is appreciated.
It analyzes the logs (Figure 10) and determi=
ne
interaction with the address PI 18=
6.46.255.168
with hour 20:30:09 indicating that the attachment was read.
Figure
10. Test 2, logs =
URL
and attached file
Prepared by:= Research Group.<= o:p>
The hit of the logs, which gives us the IP of the
target corresponds to the document of Word and the URL http://191.232.34.208/image=
s/virus.jpg.
In addition, it draws attention that, in this case, unlike the previous one=
, in
which the target account was @outlook.com, the target computer also directly
accesses the image we have linked by URL. When consulting the referential
location (Figure 11) of the address IP and possible place where the mail was
opened, it can determine the Internet provider CNT.
Figure
11. Test 2,
Referential location
=
Prepared
by: Research Group.
To deepen the study of the original message of the
received mail, just download the file “txt”
that contain the technical headers, for example:
“Delivered-To:
3hchocoatl@gmail.com
Received: by 10.176.17.71 =
with
SMTP id g7csp2940615uac;
&n=
bsp;
Sun, 30 Jul 2017 17:48:00 -0700 (PDT)
X-Received: by 10.237.36.38
with SMTP id r35mr21284604qtc.327.1501462080165;
&n=
bsp;
Sun, 30 Jul 2017 17:48:00 -0700 (PDT)”
&n=
bsp;  =
; &n=
bsp;  =
; &n=
bsp;
Tests in Tor networks
The URL http://
191.232.34.208 it will be executed from the anonymous browser to later
proceed to an analysis of the logs. As a final test of this investigation, =
the
test will be done in TOR, opening the URL in the browser. The records obtai=
ned
by the visit of the website are those shown in Figure 12:
Figure
12. Logs TOR
=
Prepared
by: Research Group.
A direction is detected IP but it is not the same =
when
performing tests on different pages, the location is not reliable and the I=
PS
is in another country despite that in the test and experiments carried out =
from
known locations.
Figure
13. Wrong locatio=
n in
networks
=
Prepared
by: Research Group.
Conclusions.
· &n=
bsp;
In the present investigation, different methods ha=
ve
been tried to obtain directions IP of a user in social networks and Webmails. Using the facebook
social network, if you try to link an image through your URL through differ=
ent
forms (publication on the wall, direct message; etc.) emphasizes in particu=
lar
that the locations of the IP addresses found correspond to servers or data
centers of Facebook. This address does not belong to the users directly and
therefore, this is not a valid method to obtain the IP address of an
objective. The same happens when performing these tests with webmails, If we try to embed in the message an image
referencing it by its URL.
· &n=
bsp;
Finally, it has been possible to verify that the
technique of embedding the image in a document of Word (using the field code
Include Picture), If it is effective for the target pursued, because once t=
hat
file is made to the recipient, Target pursued we will obtain a record of yo=
ur
address IP, as well as other data revealed by the field User-Agent, such as=
the
operating system and software used. However, it should be recalled that the
identify an IP address published where it comes from a message, it does not
guarantee the immediate location of a person; because this network can have
multiple devices that connect to that address: therefore, research just sta=
rted
with this study.
· &n=
bsp;
In the future it would be interesting to use
JavaScript, to obtain additional information about the target system (data =
such
as: computer name, user name, etc.) Another possible line would be the
investigation of IP address detection methods, when users use a proxy.
Bibliographic Ref=
erences.
Arndt, M. R., & Actis, F. J. (1996). Method of configuring a valid IP
address and detecting duplicate IP addresses in a local area network.
https://patents.google.com/patent/US5724510A/en
Arroyo J. Richard. (2016). Análisis de los del=
itos
informáticos por ataque y acceso no autorizado a sistemas
electrónicos, tipificados en los artículos 232 y 234 del
Código Orgánico Integral Penal en el Ecuador.
http://www.dspace.uce.edu.ec/bitstream/25000/5953/1/T-UCE-0013-Ab-121.pdf
Bracho, C., Fabián, C., Pupiales, C., &
Suarez, L. (2017). Auditoría de seguridad informática siguien=
do
la metodología OSSTMMv3: caso de estudio | Mask=
ana.
Maskana=
span>, 8. https://publicaciones.ucuenca.edu.ec/ojs/inde=
x.php/maskana/article/view/1471
De Juventud, E., & De, R. (2017). Jóvenes:
bullying y ciberbullying. http://www.publicacionesoficiales.boe.es
García-Maldonado, G., Joffre-Velázquez,=
V.
M., Jesús Martínez-Salazar, G., & Llanes-Castillo, A. (20=
11).
Ciberbullying: forma virtual de intimidación escolar. In Rev. Colomb. Psiquiat (Vol. 40=
, Issue 1).
http://www.scielo.org.co/pdf/rcp/v40n1/v40n1a10.pdf
González García, A. (2015). El
ciberbullying o acoso juvenil a través de Internet: un anális=
is
empírico a través del modelo del Triple Riesgo Delictivo (TRD=
).
https://www.tdx.cat/bitstream/handle/10803/384709/AGG_TESIS.pdf?sequence=3D=
1
Herrera-López, M., Romera, E. M., Ortega-Ruiz,=
R.,
& Herrera, M. (2018). Bullying y Cyberbullying en
Latinoamérica. In Revista Mexicana de Investigación Educativa
RMIE (Vol. 23). http://www.comie.org.mx/documentos/rmie/v23/n076/pdf/76005.=
pdf
Ibarra, J. D., & Electrónico, I. (2004).
Auditoría tecnológica con fines de seguridad Informáti=
ca.
Johanna Martínez Molina, K., Pacheco Meneses, =
J.,
& Zúñiga Silgado, I. (2009). FirEwal=
l-linux:
una solución dE sEg=
uridad
inFormática para py=
mEs
(pEquEñas y mEdiana=
s
EmprEsas). Revista UIS Ingenierías, 8(2).
http://www.
Martinez=
span>, F., Caicedo, J., Hernandez, R.,Caicedo, O., & Hurtado, J. (2007). Seguridad b=
asada
en parámetros SIM para entornos de comercio electrónico
móvil SIM parameter-based security
for mobile e-commerce settings. Amicus Curiae, 27(2), 56–64.
Arndt, M. R., & Actis, F. J. (1996). Method of configuring a valid IP
address and detecting duplicate IP addresses in a local area network.
https://patents.google.com/patent/US5724510A/en
Arroyo J. Richard. (2016). Análisis de los del=
itos
informáticos por ataque y acceso no autorizado a sistemas electr&oac=
ute;nicos,
tipificados en los artículos 232 y 234 del Código Orgá=
nico
Integral Penal en el Ecuador.
http://www.dspace.uce.edu.ec/bitstream/25000/5953/1/T-UCE-0013-Ab-121.pdf
Bracho, C., Fabián, C., Pupiales, C., &
Suarez, L. (2017). Auditoría de seguridad informática siguien=
do
la metodología OSSTMMv3: caso de estudio | Mask=
ana.
Maskana=
span>, 8.
https://publicaciones.ucuenca.edu.ec/ojs/index.php/maskana/article/view/147=
1
De Juventud, E., & De, R. (2017). Jóvenes:
bullying y ciberbullying. http://www.publicacionesoficiales.boe.es
García-Maldonado, G., Joffre-Velázquez,=
V.
M., Jesús Martínez-Salazar, G., & Llanes-Castillo, A. (20=
11).
Ciberbullying: forma virtual de intimidación escolar. In Rev. Colomb. Psiquiat (Vol. 40=
, Issue 1). http://www.scielo.org.co/pdf/rcp/v40n1/v40n=
1a10.pdf
González García, A. (2015). El
ciberbullying o acoso juvenil a través de Internet: un anális=
is
empírico a través del modelo del Triple Riesgo Delictivo (TRD=
).
https://www.tdx.cat/bitstream/handle/10803/384709/AGG_TESIS.pdf?sequence=3D=
1
Herrera-López, M., Romera, E. M., Ortega-Ruiz,=
R.,
& Herrera, M. (2018). Bullying y Cyberbullying en
Latinoamérica. In Revista Mexicana de Investigación Educativa
RMIE (Vol. 23). http://www.comie.org.mx/documentos/rmie/v23/n076/pdf/76005.=
pdf
Ibarra, J. D., & Electrónico, I. (2004).
Auditoría tecnológica con fines de seguridad Informáti=
ca.
Johanna Martínez Molina, K., Pacheco Meneses, =
J.,
& Zúñiga Silgado, I. (2009). FirEwal=
l-linux:
una solución dE sEg=
uridad
inFormática para py=
mEs
(pEquEñas y mEdiana=
s
EmprEsas). Revista UIS Ingenierías, 8(2).
http://www.
Martinez=
span>, F., Caicedo, J., Hernandez,
R., Caicedo, O., & Hurtado, J. (2007). Seguridad basada en
parámetros SIM para entornos de comercio electrónico mó=
;vil
SIM parameter-based securi=
ty
for mobile e-commerce settings. Amicus Curiae, 27(2),
56–64.
Óscar Fernando Castellanos, Montañez, A=
. M.
F., & Fonseca, S. L. (2007). Bases de la implementación de un mo=
delo
de inteligencia para fortalecer el desarrollotecnol&oa=
cute;gico
de la industria del software y servicios asociados en Colombia.
Ingeniería e Investigación, 27(3), 182–192.
https://revistas.unal.edu.co/index.php/ingeinv/article/view/14859
Robert T. Baum. (2003). Me=
thods
and apparatus for protecting against IP address assignments based on a false
MAC address. https://patents.google.com/patent/US7320070B2/en
Romera, E. M., Ortega-Ruiz, R., Herrera-López,=
M.,
Romera, E. M., & Ortega-Ruiz, R. (1996). Revista mexicana de investigación educativa. In Revista mexicana de
investigación educativa (Vol. 23, Issue =
76). Consejo
Mexicano de Investigación Educativa.
http://www.scielo.org.mx/scielo.php?pid=3DS1405-66662018000100125&scrip=
t=3Dsci_arttext
Sarmiento, O. P., Guerrero, F. G., & Argote, D. R.
(2016). Basic security
measures for IEEE 802.11 wireless networks. =
Ingeniería
E Investigación, 28(2), 160–167.
https://doi.org/10.1109/ITHERM.2016.7517544
Ximhai, R. (2014). Manifestaciones del Ciberbullying por
Género Entre los Estudiantes de Bachillerato. Ra Ximhai,
10(2), 235–261. http://www.redalyc.org/articulo.oa?id=3D46132726010
Zapata, L. (2012). Evaluación y mitigaci&oacut=
e;n
de ataques reales a redes ip utilizando
tecnologías de virtualización de libre distribución. <=
span
class=3DSpellE>Ingenius.Ups.Edu.Ec, 11–19.
http://ingenius.ups.edu.ec/documents/2497096/3033837/Articulo+2.pdf
PARA CITAR EL ARTÍCULO INDEXADO.=
Mor= eano Sánchez, G. V., Ávalos Yuque , E.= V., Benitez Bravo, V. H., & Beni= tez Bravo, A. G. (2020). Métodos electrónicos para obtenció= ;n de direcciones IP a través de webmail y = redes sociales. ConcienciaDigital, 3(2.= 2), 80-96. h= ttps://doi.org/10.33262/concienciadigital.v3i2.2.1247
El
artículo que se publica es de exclusiva responsabilidad de los autor=
es y
no necesariamente reflejan el pensamiento de la Revista Conciencia Digital.
El artículo qu=
eda
en propiedad de la revista y, por tanto, su publicación parcial y/o
total en otro medio tiene que ser autorizado por el director de la Revista Conciencia Digital.
[1=
] ESPOCH, CIMANT,
Riobamba, Ecuador, gabriel.moreano@espoch.edu.ec, ORCID:
https://orcid.org/0000-0002-9331-8223
[2]
Telefónica, Quito, Ecuador, vinycioavalos@hotmail.com, ORCID:
https://orcid.org/0000-0001-7458-9370
[3=
] Universidad
Tecnológica Israel, Quito, Ecuador, vhbenitez@uisrael.edu.ec, ORCID:
https://orcid.org/0000-0002-8975-3644
[4= ] Universidad Tecnológica Israel, Quito, Ecuador, agbenitez@uisrael.edu.ec, ORCID: https://orcid.org/0000-0002-8465-1059
&nb= sp; = &nb= sp; = &nb= sp; = &nb= sp; = &nb= sp; = <= /span>ISSN: 2600-5859
= &nb= sp; = &nb= sp; = &nb= sp; = &nb= sp; Vol. 3, N°2.2, p. 80-96, junio, 2020