=Paper=
{{Paper
|id=Vol-3373/paper30
|storemode=property
|title=Protection of data transmission in remote monitoring tools by anonymization
|pdfUrl=https://ceur-ws.org/Vol-3373/paper30.pdf
|volume=Vol-3373
|authors=Anatoliy Melnyk,Jean-Yves Le Boudec,Yurii Morozov,Bohdan Havano,Petro Hupalo
|dblpUrl=https://dblp.org/rec/conf/intelitsis/MelnykBMHH23
}}
==Protection of data transmission in remote monitoring tools by anonymization==
https://ceur-ws.org/Vol-3373/paper30.pdf
Protection of Data Transmission in Remote Monitoring Tools by
Anonymization
Anatoliy Melnyka,b,c, Jean-Yves Le Boudecd, Yurii Morozova,b, Bohdan Havanoa,b and Petro
Hupaloa
a
Intron ltd, Lviv, Ukraine
b
Department of Computer Engineering, Lviv Polytechnic National University, Lviv, Ukraine
c
The John Paul II Catholic University of Lublin, Lublin, Poland
d
Ecole Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland
Abstract
Data anonymization is a method of removing or encrypting personal data in a data set. At the
same time, the existing data structure is maintained for further data analysis. The purpose of
anonymization is to ensure the confidentiality of the subject's information. This paper
examines the problems of user authorization and the protection of data transmission in
remote monitoring tools. Technologies for protecting data transmission in remote monitoring
tools are selected. In the presented remote monitoring tools, data pseudonymization is used to
anonymize data. Pseudonymization of data consists in replacing the user ID with an
encrypted JWT token. Because the JWT token is encrypted using TLS, which uses revocable
symmetric keys, there is no way to detect the connection between the user and the transmitted
information during data transmission. The proposed method protects the privacy of remote
monitoring data by dividing the user authorization process into two network services and by
anonymizing the remote monitoring data. We present the design principles and the
algorithms of operation. The remote monitoring tools with data transmission protection by
anonymization were implemented as the result of the conducted research.
Keywords 1
Anonymization, data protection, protection of data transfer, remote monitoring.
1. Introduction
Remote monitoring tools obtain measurements from wireless wearable devices and process them
to support some applications such as health monitoring, localization, and crowd density estimation.
They transmit data from wireless wearable devices to cloud services and connect these services to
mobile terminals [1].
To allow interaction between users and to process the parameters of interest, users are required to
enter personal data. In comparison to technical data, the feature of personal data is that they carry
information by which you can identify the user. Therefore, it is necessary to obtain permission from
the user to process his data and store them. In most countries, the law protects personal data. For
example, in the European Union, there is a general regulation on data protection (GDPR Regulation
(EU) 2016/679) [2].
However, even if authorized by users, the collected personal data can be attacked by intruders, and
undesired privacy leaks can occur; so the task is to ensure their protection.
IntelITSIS‟2023: 4th International Workshop on Intelligent Information Technologies and Systems of Information Security, March 22–24,
2023, Khmelnytskyi, Ukraine
EMAIL: aomelnyk@gmail.com (A. Melnyk); jean-yves.leboudec@epfl.ch (J-Y. Le Boudec); yurii.v.morozov@lpnu.ua (Y. Morozov);
havano.bohdan@gmail.com (B. Havano); gypalo911@gmail.com (P. Hupalo)
ORCID: 0000-0002-8981-0530 (A. Melnyk); 0000-0003-2357-8078 (J-Y. Le Boudec); 0000-0002-3670-411X (Y. Morozov); 0000-0002-
2546-1917 (B. Havano); 0000-0003-4984-3220 (P. Hupalo)
2023 Copyright for this paper by its authors.
Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
CEUR Workshop Proceedings (CEUR-WS.org)
2. Status of the problem
All of the above raises the issue of choosing an effective mechanism for data anonymization.
As demonstrated in de-anonymization methods, i.e., the restoration of personal data from
anonymized data presented in scientific literature, removing direct identifiers such as full name and
address is insufficient to ensure confidentiality of individuals. Data set owners must be aware of the
risks of de-anonymization and apply appropriate anonymization measures before granting access to
their data sets to comply with data protection regulations. To meet this need, a procedure was defined
in [3] that informs data set owners of de-anonymization risks and helps them make decisions
regarding anonymization measures that need to be taken to comply with the General Data Protection
Regulation (GDPR). This paper demonstrates the application of the de-anonymization assessment
procedure to a customer relationship management (CRM) data set provided by a telecommunications
provider.
In addition, in [4], the authors evaluated the practical feasibility of anonymizing electronic health
records (EHR) data with regard to their suitability for making medical decisions. In a real-world
scenario, effective anonymization of data is challenging because it typically depends on the problem
and requires significant expertise in the field. As the level of data anonymity increases, the
convenience of using anonymous data decreases and most likely reaches a minimum convenience
before achieving full anonymity.
In [5], possible schemes for publishing healthcare data that guarantee more reliable data
confidentiality while preserving the usefulness of published data were proposed. Two confidentiality
models were identified: identity unlinkability and attribute unlinkability, which include the possibility
of unlinkability in data anonymization. The absence of the possibility of disconnection during data
transmission may lead to a breach of patient confidentiality, as the patient can be tracked or linked.
This prompted the authors to develop two schemes that use these confidentiality models for data
anonymization in two different data transmission scenarios.
A combination of pseudonymization and anonymization methods can provide additional protection
when data is first anonymized by removing any personal identifiers and then encrypted before storage
[6]. When data is properly anonymized by removing all identifiers, it is no longer subject to GDPR,
allowing companies to collect such data without consent and store it indefinitely.
User confidentiality is the primary criterion for allowing the transmission of confidential
information. In article [7], the Privacy Preserving Data Mining (PPDM) approach is discussed, which
analyzes data while maintaining confidentiality. PPDM methodologies are reviewed and classified
using different approaches to modify data for anonymization. In addition, a critical comparative
analysis of the advantages and disadvantages of PPDM methods is performed.
Article [8] presents a protocol for data collection with confidentiality preservation. This protocol
does not restrict the type of anonymization method and does not require a private channel. It only
requires k-anonymity to prevent attacks on confidentiality, and thus equivalent groups of owners of
data sets function as a mechanism for preserving confidentiality.
The main problem of data exchange is the intelligent control of the security of private data [9]. In
this work, a structure for evaluating data anonymization is proposed, which can evaluate typical data
anonymization algorithms in terms of the level of confidentiality preservation, data usefulness, and
productivity.
Therefore, as the analysis of recent scientific publications has shown, information systems require
user authentication to access their data, but at the same time, the data transmitted must be
anonymized. It is necessary to combine the user authentication process with the process of
anonymizing the data being transmitted. This will increase the level of data anonymity while
preserving the convenience of using anonymous data.
3. Selection of technologies for protecting the transmission of personal data
in the remote monitoring tools
Personal data is necessary for the authorization of the user in the information system. This data
must be transferred and saved. To avoid problems with processing personal data, we propose using a
remote authorization service that will process personal data. Single Sign-On technology can be used
to provide simultaneous access to various services [12].
The wireless wearable devices are connected to the user's mobile terminal through Bluetooth. The
user enters personal data into their mobile terminal. This data should not be stored in the mobile
terminal to ensure compliance with Single Sign-On technology. Therefore, the user must connect to
remote authorization services from the mobile terminal.
4. Anonymization of user’s personal data
Data anonymization is a method of removing or encrypting personal data in a data set. At the same
time, to enable data further analysis the existing data structure is supported. The goal of
anonymization is to ensure the confidentiality of the subject's information.
In accordance with the European Union's General Data Protection Regulation (GDPR), which
requires pseudonymization or anonymization of stored information of persons living in the EU,
anonymized data are not classified as personal data and are therefore not subject to the rules of this
regulation. It allows organizations to use anonymized data without violating the rights of data
subjects. Similarly, data anonymization is a core component of HIPAA requirements, a US regulatory
act that governs the use of private health information (PHI) in the healthcare industry [13].
First, information that allows the identification of a person is anonymized. This includes name, as
the most important identifier in the data set, credit card details, mobile numbers, photos, passwords,
security questions, health data, and more.
Anonymization is used to preserve the confidentiality of the patient's information about his illness,
to preserve the confidentiality of customers whose data is used in digital advertising in marketing
social networks, to preserve the confidentiality of information about company employees, which is
collected for the purpose of increasing productivity, optimizing work, and improving security.
The following methods are used to anonymize data [14,15,16,17]:
Data masking by providing access to a modified version of sensitive data.
Pseudonymization of data by replacing private identifiers with pseudonyms or fake
identifiers.
Generalization of data by excluding certain data from it to make it smaller identified.
Shuffle or rearrange the data by changing the value of the attributes of the dataset.
Data perturbation using rounding and random noise techniques.
Creating synthetic data using pattern-based mathematical systems or features in the original
data set.
Data in remote monitoring tools are divided into personal data and technical data. Personal data is
used infrequently, so it can be separated from technical data and protected by reliable cryptographic
means. Technical data in this case is classified as big data. The processing of technical data requires
significant computing resources both in the mobile terminal and in the cloud service. Therefore, from
a productivity point of view, it is advisable to carry out minimal protection of technical data or not to
protect it at all if it can be anonymized [10].
Personal data and technical data should be linked with some user identifier. Each time the mobile
terminal and cloud service interact, the user identifier is transmitted together with the technical data.
Technical data, as well as personal data, are interesting to the attacker in case they can establish
the identity of a certain user. Data protection can be done using user anonymization technology, such
as shown in [19] and [18]. However, these technologies are complex and require significant
computing resources.
An alternative is to break the connection between personal and technical data, using the method in
[21]. With this method, the mobile terminal establishes a connection with the cloud service, transmits
the encrypted user identifier and date/time label, and obtains a temporary “authorization user id”. The
cloud service decrypts the received identifier and date/timestamp and compares this label with the
running time. If time is significantly different, then the data is falsified. The authorization user id is
changed periodically. With every interaction between the mobile terminal and the cloud service,
another authorization user id accompanies the technical data. This does not allow the establishment of
a direct connection between technical data and the user.
4.1. User authentication
Authentication is the process of secure user identification. The authentication mechanism provides
access control for systems by verifying that the user credentials match the data in the authorized user
database or on the data authentication server. Authentication methods are divided depending on the
type of resource, structure and method of network organization, object remoteness and technology
used in the recognition process.
Remote monitoring tools operate using a client-server architecture. User authentication takes place
in the cloud service, and the mobile terminal contains the user interface for this service. It is advisable
to divide the processes of authentication and remote monitoring between two cloud services. One of
these services deals only with authentication and informs the other service about the authenticated
user.
The most reliable and rich cloud authentication services use oAuth2 technology [20]. It allows
applications to exercise limited access to user accounts on HTTP services. It uses an authentication
token sent to the user, allowing a third-party application to access the user's account.
OAuth2 is a widely used security standard that provides secure access to protected resources in a
way convenient for the web API. As it uses HTTPS and sends the authentication token in headers, this
protocol can be used in web, desktop, and mobile applications.
Given that the technology is decentralized, it does not matter how it will be used. It works in
parallel and independently. Therefore, you can use several authorization services at the same time,
depending on your preferences. Accordingly, the remote monitoring cloud service receives
information about the authorized user of any of the cloud services built on this technology.
4.2. Protection of data transmission in remote monitoring tools
Data on the measuring device is not tied to the user. They are transmitted to the user's mobile
terminal using Bluetooth technology, which provides protection against falsification and modification
of transmitted data. On the other hand, energy efficiency requirements do not recommend the use of
additional means of information security.
In a mobile terminal, data is associated with the user and becomes sensitive to interference. At the
same time, the requirements for energy efficiency are not as strict as for metering devices. Therefore,
it is advisable to apply the protection of the communication channel. Use TLS 1.3 or IPsec to protect
transmitted data. Most IPsec implementations are more reliable than some TLS 1.3 implementations,
but more energy intensive. Therefore, the TLS 1.3 protocol is most often used in mobile terminals
and, accordingly, in cloud services [23].
This technology is used to make sure that any data transmitted between the mobile terminal and
the cloud service remains impossible to read [22]. Encryption algorithms are used to encrypt data
during transmission, not allowing hackers to read it when they are sent. This information may be
confidential or personal, which may include technical or personal data.
JSON Web Token (JWT) is an open standard (RFC 7519) that defines a compact and autonomous
way of safely transmitting information between parties as a JSON object [23]. JWT is a base64
encoded JSON object that is considered a secure way of transmitting information between two
participants.
Improper storage of tokens leads to their potential theft and malicious actions.
Most often, JWT is used for authorization. Once the user is logged in, each subsequent request
contains JWT, which allows the user to access the routes, services and resources allowed by this
token. Authorization is a process that today makes extensive use of JWT through small resources and
the ability to use across domains.
Compared to other standards such as Simple Web Tokens (SWT) and Security Statement Markup
Language Tokens (SAML), JWT is more compact and more appropriate to use for different platforms,
especially mobile terminals.
4.3. Base request structure
A typical request with technical data consists of an authorization header and payload as the request
body. Figure 1 shows an example of data structure in the request.
POST /echo/post/json HTTP/1.1
Host: cloud-service.com
Authorization: Bearer JWT-token-value
Content-Type: application/json
Body:
{ "spo2_values": [],
"pulse_values": []
}
Figure 1. An example of data structure in the request
The authorization header has the JWT access token, received from auth service. The payload is a
set of technical data without any personal data in it. The JWT token is used to process data and store it
in relation to a specific user in the database. Since personal data are missing, a user identifier from
auth service is the only possible way to map technical data to a specific user.
Once a request from the mobile terminal goes to cloud service, this request must be authorized
first. A JWT token from the authorization header is taken. The next step is to verify this token using a
public key in JWKS format. If the token is verified and valid it is possible to extract some data from
it.
Figure 2 shows the general structure of a decoded token.
{
iss: 'issuer',
sub: 'auth-user-id',
aud: [],
iat: 1589809948,
exp: 1589896348,
azp: 'app_ID',
scope: „list of scopes‟
}
Figure 2. General structure of the decoded token
It is also possible to add another authorization layer using allowed scopes which are also defined
in the access token. Using sub value from the token object we can identify an anonymized user by id.
So, even without any personal data, it is possible for an information system to transfer, process and
store technical data in relation to some specific user in the database.
4.4. Protection of data transfer via VPN
Anonymization protects data from identification but does not protect against modification and
unauthorized viewing. Virtual private network (VPN) technology is used to protect the transmitted
data. It consists in creating cryptographically protected tunnels through open communication
networks. The essence of tunnelling is that anyone outside (for example, a cryptanalyst) does not have
access to the information being transmitted.
TLS 1.3 or IPsec protocols are used to create a VPN. As it was mentioned above, most IPsec
implementations are more reliable than some TLS 1.3 implementations, but more energy intensive.
Therefore, the TLS 1.3 protocol is most often used in mobile terminals and, accordingly, in cloud
services [23, 26].
TLS 1.3 (Transport Layer Security) is a standard technology for securing Internet connections and
protecting any sensitive data that is transmitted between two systems by preventing cryptanalysts
from reading and modifying any information transmitted. The handshake protocol is responsible for
creating a tunnel in TLS 1.3. This protocol uses public key technology in the form of X.509 digital
certificates.
Figure 3 shows what happens during the TLS 1.3 handshake.
Figure 3. TLS 1.3 handshake protocol
The sequence of interaction steps is as follows:
1. Mobile terminal (MT) sends a request to the server for a secure session.
2. The cloud service (CS) responds by sending its X.509 digital certificate to the client.
3. Mobile terminal obtains the X.509 digital certificate of the server and authenticates the server
using a list of known certification authorities.
4. Mobile terminal generates a random symmetric key and encrypts it with the public key of the
cloud service.
5. Mobile terminal and cloud service now know the symmetric key and can use it to encrypt and
decrypt the information contained in the request of the mobile terminal and the response of the
cloud service.
The symmetric key obtained using the handshake protocol is used to encrypt all transmitted
information. This information can be confidential or non-confidential, which may include biometric
data. The lifetime of a symmetric key is limited so as not to give the cryptanalyst enough data to
analyze.
This technology is used to make sure that any data transferred between the mobile terminal and the
cloud service remains unreadable. Encryption algorithms are used to encrypt data in transit,
preventing cryptanalysts from reading it as it is sent. This information can be of any kind, which may
include biometric data.
5. Structure and operation of remote monitoring tools with data
transmission protection
The proposed structure of remote monitoring tools with technical protection of data transmission is
shown in Figure 4. The remote monitoring tools include the following components:
measuring device,
mobile terminal,
authentication service,
cloud service,
database.
The measuring device consists of sensors that collect information and transmit it to a mobile
terminal.
The mobile terminal consists of the sensors data receiver for data acquisition from sensors of
measuring devices, a data transmission unit for data transmission to the cloud service (for interaction
with the cloud service), and an authentication unit (for interaction with the authentication service).
Measuring
Device
Figure 4. The structure of remote monitoring tools
In the presented remote monitoring tools, data pseudonymization is used to anonymize data.
Pseudonymization of data consists in replacing the user ID with an encrypted JWT token. Because the
JWT token is encrypted using the TLS 1.3 protocol, which uses revocable symmetric keys, there is no
way to detect the connection between the user and the transmitted information during data
transmission.
The Authentication Service includes decentralized user authentication technologies that operate
independently of each other.
The cloud service receives data from the mobile terminal as a REST API. As in the mobile
terminal, there is an authentication unit. The business logic block is an important part of the cloud
service, which is responsible for the processing and storage of information.
The database is a repository where personal and technical data that has been processed in the
business logic block is stored.
The processing and storage of user personal data in the system are absent. Personal data, such as
login and password, are used for authentication. They are not saved anywhere in the information
system but passed from the mobile terminal to auth service for further authorization. The cloud
service never receives or stores any personal data.
The principles of mobile terminal (MT) operation with the protection of technical data
transmission are shown in Figure 5.
Figure 5. The flow of operation of the client application on the mobile terminal
The very first step is to authenticate the user. The client application for the mobile terminal first
checks the presence of an up-to-date authentication access token in the local secret storage. If the
token is missing or the expiration time has occurred, the application will require the user to
authenticate to the authentication server. After successful authentication, the application switches to
continuous data transfer of the parameters of interest to the cloud service. MT sends only technical
data to the cloud service along with a JWT token without any personal data. The JWT token only
knows the authentication user id, which is used to map data in the cloud service DB. Note that it is not
possible to impersonate users using this identifier, thus every user is anonymized in the information
system.
If the measuring device contains means of accessing the Internet, for example, WiFi, then the
principle of operation in this form is shown in Figure 6.
Figure 6. The flow of operation of the client application on the measuring device
Like the previous algorithm, the first step is user authentication. The Measuring device app first
checks for a valid authentication access token in the local secret store. If the token is missing or
expired, the application will require the user to authenticate with the authentication server. But, in this
case, the Measuring device app will first receive the Device Code from the authentication server,
which is transmitted by the mobile terminal. On the mobile terminal, the user enters a login and
password, because there is no keyboard on the measuring device. After successful authentication, the
Measuring device app receives a JWT token and switches to the constant transfer of the data of the
parameters of interest to the cloud service. The measuring device sends only technical data to the
cloud service along with the JWT token without any personal data.
In this way, the transmitted data is completely anonymized in the information system.
The algorithm of operation of Cloud Service (CS) for processing technical data is shown in Figure
7.
Figure 7. The flow of operation of CS for protecting the transmission of technical data
Once MT obtains the Access token it is possible to make authorized requests to the Cloud Service.
Each request from MT to CS should have an authorization header with a JWT token obtained from
auth service.
Auth service uses RSA Signature with SHA-256 for signing tokens. Since it uses a private/public
keypair, it verifies the tokens against the public key for CS. The public key is in the JSON Web Key
Set (JWKS) format.
CS should obtain JWKS with the public key to be able to verify the access token received from
MT and to authorize the request and the user. It is important that CS doesn‟t make any requests to
obtain any personal data. It means that the developed information system does not have any personal
data transmission or storage.
Using JWKS, CS can authorize the request and process incoming data if authorization is
successful. In other cases, MT will receive an unauthorized response. During data processing, CS
retrieves the user identifier from JWT and links processed data in storage with the authentication user
id. the last step is to send the response from CS to MT.
6. Device authentication process
With input-constrained devices that connect to the internet, rather than authenticate the user
directly, the device asks the user to go to a link on their smartphone and authorize the device. This
avoids a poor user experience for devices that do not have an easy way to enter text. To do this,
device apps use the Device Authorization Flow (ratified in OAuth 2.0) [27], in which they pass along
their Client ID to initiate the authorization process and get a JWT token. Flow example is presented
on figure 8.
Figure 8. Device app screen (left), smartphone authorization screen (middle) and device app after
successful authorization (left).
7. Conclusion
Personal data often becomes the object of attack by intruders, so the task is to ensure their
protection. Personal data in remote monitoring tools should not be stored in users' mobile terminals,
cloud services, but should be transferred between them. Therefore, it is necessary to protect this data
in all these places. That is, it is necessary to provide means of secure authorization, both in mobile
terminals and in cloud services, and to secure the transmission of this data.
A literature review has revealed that there is a problem with deanonymization of data, as well as
the depth of anonymization that is related to the convenience of data usage. This is linked to the
problematic non-binding of identity. However, the combination of pseudonymization and
anonymization methods can provide additional protection by removing any personal identifiers. The
main issue in data exchange is the intelligent control of private data security.
Based on this, it is proposed to divide data into personal and technical data. At the same time,
personal data are protected as much as possible, and technical data are anonymized. In measuring
devices, basic protection of technical data is achieved by using Bluetooth technology. No personal
data is stored in the mobile terminal. User authentication is carried out on decentralized cloud services
using oAuth2 technology. From the mobile terminal, information is transmitted to the cloud service
via HTTPS (TLS) using JWT tokens to authorize requests.
The proposed method to protect the transmission of remote monitoring data is to divide the user
authorization process into two network services and the subsequent anonymization of remote
monitoring data. Thus, the data collection and processing service only receives anonymous data.
Based on this method, the principles of design, algorithms of operation and the remote monitoring
tools with the protection of data transmission have been developed.
In the future, it is planned to expand the research not only in terms of data transmission but also
data storage. Anonymization of stored data is a separate task, the solution of which will allow for a
comprehensive solution to the issue of personal data protection in data collection and processing
systems.
8. References
[1] A. Melnyk, Y. Morozov, B. Havano, P. Hupalo. HealthSupervisor: Mobile Application for Round-
the-Clock Remote Monitoring of the Human Functional State (keynote). Proceedings of the 2nd
International Workshop on Intelligent Information Technologies & Systems of Information
Security with CEUR-WS. Khmelnytskyi, Vol-2853, Ukraine, March 24–26, 2021, pp. 24-37.
[2] Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the
protection of natural persons with regard to the processing of personal data and on the free
movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation)
(Text with EEA relevance), [2016] OJ L 119/1.
[3] A. Bampoulidis, A. Bruni, I. Markopoulos, M. Lupu, Practice and Challenges of Anonymisation
for Data Sharing. In: Dalpiaz, F., Zdravkovic, J., Loucopoulos, P. (eds) Research Challenges in
Information Science. RCIS 2020. Lecture Notes in Business Information Processing, vol 385.
Springer, Cham. doi:https://doi.org/10.1007/978-3-030-50316-1_32.
[4] Z. Zuo, M. Watson, D. Budgen, R. Hall, C. Kennelly, N. Al Moubayed Data Anonymization for
Pervasive Health Care: Systematic Literature Mapping Study JMIR Med Inform 2021 doi:
10.2196/29871
[5] KM Chong, A. Malip, Bridging unlinkability and data utility: Privacy preserving data publication
schemes for healthcare informatics, Computer Communications, 191 (2022) 194-207,
doi:https://doi.org/10.1016/j.comcom.2022.04.032.
[6] G.M.S. Ross, Y. Zhao, A.J. Bosman, A. Geballa-Koukoula, H. Zhou, C.T. Elliott, M.W.F. Nielen,
K. Rafferty, G.IJ. Salentijn, Best practices and current implementation of emerging smartphone-
based (bio)sensors - Part 1: Data handling and ethics,TrAC Trends in Analytical Chemistry, 158
(2023) doi: https://doi.org/10.1016/j.trac.2022.116863.
[7] M. Binjubeir, A. A. Ahmed, M. A. B. Ismail, A. S. Sadiq and M. Khurram Khan, Comprehensive
Survey on Big Data Privacy Protection, in IEEE Access, 8 (2020) 20067-20079. doi:
10.1109/ACCESS.2019.2962368.
[8] S. Kim, Y. Dohn Chung, An anonymization protocol for continuous and dynamic privacy-
preserving data collection, Future Generation Computer Systems, 93 (2019) 1065-1073,
doi:https://doi.org/10.1016/j.future.2017.09.009.
[9] C. Ni, L. Shan Cang, P. Gope, G. Min, Data anonymization evaluation for big data and IoT
environment, Information Sciences, 605 (2022) 381-392,
doi:https://doi.org/10.1016/j.ins.2022.05.040.
[10] G. Zhao, D. Zheng and K. Chen, Design of single sign-on, IEEE International Conference on E-
Commerce Technology for Dynamic E-Business, 2004, pp. 253-256, doi: 10.1109/CEC-
EAST.2004.34.
[11] A. Jayanthilladevi, K. Sangeetha and E. Balamurugan, Healthcare Biometrics Security and
Regulations: Biometrics Data Security and Regulations Governing PHI and HIPAA Act for Patient
Privacy, 2020 International Conference on Emerging Smart Computing and Informatics (ESCI),
Pune, India, 2020, pp. 244-247, doi: 10.1109/ESCI48226.2020.9167635.
[12] A. Gkoulalas-Divanis, G. Loukides and J. Sun, "Toward smarter healthcare: Anonymizing medical
data to support research studies," in IBM Journal of Research and Development, vol. 58, no. 1, pp.
9:1-9:11, Jan.-Feb. 2014, doi: 10.1147/JRD.2013.2288173.
[13] D. Gunawan, Y. S. Nugroho, Maryam and F. Y. Al Irsyadi, Anonymizing Prescription Data
Against Individual Privacy Breach in Healthcare Database, 2021 9th International Conference on
Information and Communication Technology (ICoICT), Yogyakarta, Indonesia, 2021, pp. 138-143,
doi: 10.1109/ICoICT52021.2021.9527430.
[14] B. Ouafae, R. Mariam, L. Oumaima and L. Abdelouahid, Data Anonymization in Social Networks
State of the Art, Exposure of Shortcomings and Discussion of New Innovations, 2020 1st
International Conference on Innovative Research in Applied Science, Engineering and Technology
(IRASET), Meknes, Morocco, 2020, pp. 1-10, doi: 10.1109/IRASET48871.2020.9092064.
[15] S. Dimopoulou, C. Symvoulidis, K. Koutsoukos, A. Kiourtis, A. Mavrogiorgou and D. Kyriazis,
Mobile Anonymization and Pseudonymization of Structured Health Data for Research, 2022
Seventh International Conference On Mobile And Secure Services (MobiSecServ), Gainesville,
FL, USA, 2022, pp. 1-6, doi: 10.1109/MobiSecServ50855.2022.9727206.
[16] C. Stergiou, KE. Psannis, Efficient and secure BIG data delivery in Cloud Computing. Multimed
Tools Appl 76, (2017) 22803–22822, doi: 10.1007/s11042-017-4590-4.
[17] N. Elanshekhar and R. Shedge, An effective anonymization technique of big data using suppression
slicing method, 2017 International Conference on Energy, Communication, Data Analytics and
Soft Computing (ICECDS), 2017, pp. 2500-2504, doi: 10.1109/ICECDS.2017.8389902.
[18] A. Salami, J. Andreu-Perez and H. Gillmeister, "Towards Decoding of Depersonalisation Disorder
Using EEG: A Time Series Analysis Using CDTW," 2020 IEEE Symposium Series on
Computational Intelligence (SSCI), 2020, pp. 548-553, doi: 10.1109/SSCI47803.2020.9308273.
[19] A. Melnyk, Y. Morozov, B. Havano and P. Hupalo, Protection of Biometric Data Transmission and
Storage in the Human State Remote Monitoring Tools, 2021 11th IEEE International Conference
on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications
(IDAACS), 2021, pp. 301-306, doi: 10.1109/IDAACS53288.2021.9661047.
[20] OAuth 2.0, May 2021, URL: https://oauth.net/2/
[21] D. Wagner and B. Schneier, Analysis of the SSL 3.0 protocol, The Second USENIX Workshop on
Electronic Commerce Proceedings, vol. 1, no. 1, 1996, pp. 29-40.
[22] The SSL/TLS Handshake: an Overview, May 2021, URL: https://www.ssl.com/article/ssl-tls-
handshake-overview/
[23] Jones et al. JSON Web Token (JWT). RFC 7519. RFC Editor, May 2015. URL: http://www.rfc-
editor.org/rfc/rfc7519.txt.
[24] Y. Cheng, W. Kang M. Xiao, Model checking of SSL 3.0 protocol based on SPIN, 2010 2nd
International Conference on Industrial and Information Systems, Dalian, 2010, pp. 401-403, doi:
10.1109/INDUSIS.2010.5565737.
[25] Denniss, et al. OAuth 2.0 Device Authorization Grant. RFC 8628. RFC Editor, August 2019. URL:
https://www.rfc-editor.org/rfc/rfc8628.