Hindawi

Security and Communication Networks

Volume 2022, Article ID 5300253, 14 pages
https://doi.org/10.1155/2022/5300253

Research Article
A Secure IoT and Cloud Computing-Enabled e-Health
Management System

Chanapha Butpheng ,1 Kuo-Hui Yeh ,1,2 and Jia-Li Hou1

1
Department of Information Management, National Dong Hwa University, Hualien 97401, Taiwan
2
Department of Computer Science and Engineering, National Sun Yat-sen University, Kaohsiung 804, Taiwan

Correspondence should be addressed to Kuo-Hui Yeh; [email protected]

Received 3 November 2021; Revised 19 April 2022; Accepted 11 May 2022; Published 2 June 2022

Academic Editor: Azees M

Copyright © 2022 Chanapha Butpheng et al. ,is is an open access article distributed under the Creative Commons Attribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is
properly cited.
Utilization of the Internet of ,ings (IoT) technology is virtually ubiquitous across various life disciplines. When the IoT is
integrated into e-health system to enable more real-time on-demand services, it brings significant convenience to the physicians
and patients. However, the risk of encountering unreliable information and potential security threat is promptly raised at the same
time. ,erefore, in this research, we establish a secure IoT and cloud computing-enabled e-health management system in which a
distinctive authentication scheme is adopted. ,e proposed system ensures all major security requirements, such as confi-
dentiality, entity anonymity, repudiation tracking, data integrity, and innate resistance to man-in-the-middle, location spoofing,
and replay attacks. Moreover, our system can effectively be executed on low-powered processing units and simultaneously
guarantees greater security than historically available alternatives.

1. Introduction drivers of this trend. However, all new technology has

potential security vulnerabilities that emerge as the tech-
,e adoption of smart object technology has rapidly driven a nology matures. Every IoT object may be a potential threat
huge advancement in wireless sensor-based application vector in the sense that each IoT device represents a possible
development. ,ese applications deliver efficiency of data vulnerable entry point that can be exploited by malicious
transfer, and their nature of ubiquity enables real-time and actors. ,ere are two security aspects we may consider: (a)
on-demand Internet of ,ings- (IoT-) based services into the assurance of robust physical protection of IoT devices;
daily life. ,e most important variable to consider when (b) the assurance of data availability, confidentiality, in-
developing novel IoT-based applications is security during tegrity, and privacy while engaging in normal operation.
data processing. Potential threats constantly emerge as Given the advanced nature and rapid innovation of IoT
technology evolves. Hello Barbie is an example of a con- technologies, there is a great demand and expectation that
sumer product with IoT functionality that could be utilized revolutionary security solutions will be adapted specifically
as a possible threat vector by bad actors to spy on consumers to IoT-based objects and their specific applications. ,e
utilizing the built-in camera and voice-interaction capa- demand stems from the shortcomings of traditional security
bilities. In addition, hackers might be able to use network- protocols which are not designed for IoT devices. Firewalls
monitoring-driven devices to collect personal sensitive in- that integrate protocols to manage network traffic at the
formation from network attached IoT devices. application level are an example of a solution that is capable
IoT continued to emerge as one of the fastest accumu- of regulating high-level Internet traffic. Nevertheless, fire-
lations of communication standards by which smart objects walls are not a panacea in regard to end-point IoT devices.
used in our daily lives are integrated with the Internet. ,e Moreover, IoT objects are frequently mission-specific and
onboard computing and communication capabilities are the possess limited resources to accomplish their respective
 2037, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1155/2022/5300253, Wiley Online Library on [24/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
2 Security and Communication Networks

missions. Hence, traditional security solutions must be elements necessary for e-health systems that utilize IoT-
restructured for the specific security requirements of IoT cloud technology. ,en, section 3 introduces our proposed
devices to secure IoT-based applications and systems as a system with an authentication protocol, and section 4 ex-
whole. For instance, Yousefnezhad et al. [1] defined fun- amines the robustness of our proposed system. Next, section
damental security elements: data availability, confidentiality, 5 evaluates the effectiveness of the proposed system through
and integrity. Moreover, they outlined a lifecycle security a proof-of-concept implementation. Lastly, the conclusion
solution comprised of three stages: beginning, middle, and and suggestions for future research can be found in section 6.
end of life stages. ,us, the required privacy and security
elements have been classified utilizing the concept of IoT 2. Related Works
network architecture. More concretely, the authors devel-
oped a security architecture which categorizes the primary ,is section summarizes security as it is today and outlines
security requirements for the fundamental layer, the per- fundamental security elements necessary for e-health sys-
ceptual layer, the network layer, the support layer, and the tems that utilize IoT-cloud technology.
application layer. ,e security requirements for each layer
are agreement of encryption/key, verification of identifica-
tion, protected cloud computing, robust antivirus abilities, 2.1. -e Current State of the Art. ,e primary security pri-
authentication verification, and data privacy assurance. orities for e-health systems that utilize IoT-cloud technology
Furthermore, Hathaliya and Tanwar [2] presented the im- are robust preservation of security/privacy in regard to
portance of maintaining privacy/security within a e-health patient health information. Hence, many systems, inte-
system and an extensive summary of privacy/security so- grating privacy policies and multilevel security mechanisms,
lutions. In regard to data communication, with the popu- have been proposed to be more capable of managing patient
larity of IoT applications, the Internet is utilized to health information. Recently, Butpheng et al. [4] presented a
communicate a huge amount of generated data. ,is method practical amalgamation of cloud computing and IoT-tech-
of communication is vulnerable to attack from bad actors. nologies in an e-health setting. ,e authors demonstrated
An IoT device may be most vulnerable when communi- that integration/implementation of cloud computing and
cating, storing, or processing data. Gardašević et al. [3] IoT-technologies (CIoT) is achievable in an efficient man-
identified security and privacy threats/attacks for IoT smart ner. ,is integration improves health diagnostics, treatment
healthcare systems, such as jamming, eavesdropping, timeliness, and health outcomes and reduces costs. CIoT is
spoofing, man-in-the-middle attacks, and so on. In order to comprised of interconnected smart hardware, custom sys-
secure the system against these attacks, a solution has been tems, and custom applications that communicate over the
developed to support the advancement and widespread Internet. ,e authors provided a comprehensive technical
utilization of distributed computing. architecture with an inclusive list of privacy/security re-
Typically, security is implemented and aligned with the quirements for e-health systems that utilize IoT-cloud
target system’s operation process, including device authen- technology including identification, authentication, and
tication, firewall integration, secure booting, efficient network authorization. Aligning with the proposed architecture,
management, access control, and updates/patches. Secure security solutions can be effectively designed to decrease the
interdevice communication/authentication offers significant probability of an attack via examining and identifying
security benefit for IoT crucial elements. On the other hand, vulnerabilities from each security aspect.
the academic community has exhaustively dedicated time and Similarly, Wang and Cai [5] proposed a secure com-
resources to utilize cloud computing to solve IoT efficiency munication protocol for incorporating healthcare data and
and scalability issues. ,e cloud computing model is an Named Data Networking (NDN) based on IoT technology
extremely powerful tool because of its ability to be quickly and Edge-cloud computing (SHNIE) to secure and expedite
implemented and scaled through the use of virtualized re- the delivery of medical data and reduce latency and cost.
sources management. ,e utilization of IoT devices and its SHNIE can efficiently deliver aggregated NDN-medical data
amalgamation with cloud computing provides network to multiple users from the nearest edge devices that own the
benefits, including communication efficiency, quick access, required medical data. ,e security of communications is
and intelligent interconnection capabilities, and provides realized through a hash ciphertext of the provider’s name
complementary capabilities when integrated as a component and IDs as a verification token in the NDN to preserve
in an efficient system that is scalable and flexible. ,erefore, privacy and security. Moreover, Ray et al. [6] designed a
our study proposes a secure e-health management system novel solution to deal with issues of interoperability, het-
operated with IoT and cloud computing. We begin by de- erogeneity, and Internet-aware resistances. Proposed devices
fining fundamental required security elements for an e-health will collect patient data instantaneously and then process
system that utilizes IoT-cloud resources. ,en, we will in- and analyze it with embedded Internet connected elec-
troduce a robust authentication mechanism that operates tronics. ,e proposed devices utilize the Internet for all
within Body Sensor Networks (BSN) technology to ensure interactions and communication and are capable of remote
that the fundamental security elements, including attack monitoring and controlling. ,en, the authors further
resistance and data integrity/confidentiality, are satisfied. propose an e-healthcare architecture that incorporates
Our work is structured accordingly: section 2 summa- blockchain technology based on IoT that is classified as
rizes security as it is today and outlines fundamental security follows: IoT e-healthcare; blockchain platform; connectivity;
 2037, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1155/2022/5300253, Wiley Online Library on [24/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Security and Communication Networks 3

and IoT devices. Furthermore, Shewale and Sankpal [7] such as user convenience and protocol efficiency, for more than
utilized various medical technologies such as real-time 20 years. In addition, secure communication requires anon-
monitoring, patient data administration, and healthcare ymous and unpredictable identities and robust session keys
supervision via BSN technology to observe patients’ must be predetermined to ensure secure communication be-
heartbeats and blood pressure. ,ey introduced a moni- tween objects. Moreover, simple authentication and login
toring system to detect the status of each patient’s health without the use of session keys is easily vulnerable to security
data, collect it, then forward the data to the server using Wi- threats. Regardless of claims that various security schemes such
Fi module based wireless communication. In 2019, Chen- as Secure Sockets Layer (SSL) and Transport Layer Security
thara et al. [8] identified essential privacy/security elements (TLS) can be utilized to ensure that security remains strong
for e-health systems that utilize cloud computing. ,ey after authentication, they are ultimately inefficient due to
engineered a strengthened security and privacy system to exorbitant computational cost. ,erefore, a solution with
manage numerous types of vulnerable/private e-health data session key agreement scheme is essential to guarantee robust
including user/patient data, diagnostic data, and treatment and secure communication. For the purpose of this study, our
data. Next, Koutli et al. [9] outlined a VICINITY security team attempts to demonstrate a protected communication
framework that can be integrated into e-health applications protocol for e-health systems that utilize IoT-cloud technology
that utilize IoT devices to service elderly patients. In the for IoT-cloud-based e-health systems. However, e-health
scheme, they demonstrated how Ambient-Assisted-Living systems that utilize IoT-cloud technology always rely on
(AAL) and mHealth (VICINITY IoT platform) can be used various technologies such as sensors, actuators, and commu-
in conjunction to satisfy all required security/privacy ele- nication modules for data acquisition, manipulation, trans-
ments. ,eir e-health applications are allowed for the secure mission, real-time data analysis, data storage, and application
processing of personal medical data and remote data storage. functions. Hence, it is essential to develop robust protocols
In 2018, Chattopadhyay et al. [10] coined the term In- being perfectly capable of operational conditions for IoT
ternet of Medical ,ings (IoMT) since IoT object adoption in medical devices and achieving the required security as well.
healthcare has become ubiquitous. IoMT provides an un-
paralleled level of patient/doctor accessibility through the 2.2.2. Data Encryption. As suggested by Yeh [13] and Yeh
utilization of instantaneous management of patient data,
[14], the one-way hash operations maintain qualified security
treatment support, data processing, and other services. ,e
in an effective manner, while the exclusive-or protocol may be
authors utilized a cryptosystem to safeguard communication
vulnerable to attackers. As exclusive-or operation is capable of
and authentication between local processing units (LPU),
resisting ciphertext only attacks, these protocols are the min-
smart sensors, and transmission gateways. ,e proposed imum acceptable level of cryptanalytic security. As a result, the
cryptosystem is integrated with a BSN comprising mainly of integration of exclusive-or operation must be given careful
wearable and pervasive IoTdevices as a healthcare monitoring consideration when designing security protocols. For example,
system. Alihamidi et al. [11] proposed a Blockchain archi- publicly transmitted text must utilize volatile cipher formulae
tecture enabling fog computing technology to secure e-health in conjunction with exclusive-or procedures. Moreover, these
systems. ,ey designed a three-layered IoT-based solution for two constructs must not simply and directly be applied to the
e-Health systems with a Wireless Sensor Network (WSN). cipher. In addition, Ding et al. [15] determined that in order to
,ese three major classes are differentiated by their pro-
protect network communication, data encryption must be
grammed level of access and connectivity between the con-
utilized. Data encryption technology can deliver security within
nected system units. Various vectors of access can be used for
a computer network and its communication. Data encryption
evaluating the security effectiveness of any IoT architecture.
is utilized for the transformation of data based on pre-
Rahmani et al. [12] applied fog-computing to e-health systems determined bespoke rules/algorithms. ,is enables secure data
that utilize IoT-cloud technology by introducing a geo- transmission and the data can be decrypted at the destination
graphically distributed intermediate intelligence layer in be- through the use of unique unlocking methods. ,en, Al-Haija
tween the cloud devices and the sensor devices. ,eir solution et al. [16] suggested the use of flow cryptography to secure
addresses problems with agility, expandability, and depend- information between communicated parties through the use of
ability. ,e prototype, called a smart e-health gateway health encryption to transform plaintext into ciphertext. Subse-
monitoring system, was found effective at enhancing system quently, a designated key can be utilized to decipher the ci-
intelligence, energy efficiency, mobility, performance, inter-
phertext into plaintext. ,e following is a list of standard data
operability of security protocols, and reliability.
encryption algorithms utilized to secure network traffic: Data
Encryption Standard (DES) algorithm, MD5 algorithm, and
2.2. Security Requirement for IOT-Cloud-Based E-Health Rivest–Shamir–Adelman (RSA) algorithm [15, 16].
Systems. In section B, we introduce required security ele-
ments for e-health systems that utilize IoT-cloud technology
management frameworks. ,e following are the major se- 2.2.3. Resistance against Spoofing Attacks.
curity requirements. IoT-cloud-based communication architectures are primarily
built atop a variant of traditional wireless sensor networks,
2.2.1. Access Control: Required for Secure Communication. called body sensor networks (BSN), consisting of many
Researchers have thoroughly examined dynamic identity- different body biosensors. ,e primary problem to address is
based authentication protocols and identified their advantages, the security of these individual biosensors considering that
 2037, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1155/2022/5300253, Wiley Online Library on [24/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
4 Security and Communication Networks

they collect and retain private health-related biodata. Hence, attackers. Firstly, in order to defend the system from
our proposed secure e-health system that utilizes IoT-cloud spoofing and the unauthorized access of data, a mutual
technology should carefully consider several different se- authentication is utilized between commination devices.
curity requirements such as user identification, secure Secondly, the system must achieve and maintain anonymity
network switching, data anonymity/untraceability, and and remain untraceable to ensure that data collected by
spoofing attack resistance at both the device and system biosensors, private personal data, or patient diagnostic data
level. ,ese required security elements could be addressed is not disclosed. ,ird, the system must be hardened against
via an anonymous authentication protocol. Stepien et al. [17] forgery and replay attacks. In addition, in study [2], the
claimed that the most efficient method to secure IoT-based authors presented an extensive security/privacy framework
systems is to embed security solutions into the IoT devices for Healthcare 4.0, which established an integrated block-
directly. ,e best practice is that the IoT devices should be chain solution for different healthcare applications. ,eir
capable of accommodating and maintaining authenticity. framework improved patient diagnostics and ensured se-
,at is, IoT-based systems must be capable of preventing curity/privacy of collected healthcare information.
unauthorized access while allowing flexible interoperability
between connected devices under ad hoc network condi- 3. Proposed IOT-Cloud-Based E-Health System
tions. Moreover, to safeguard the data security at the IoT
device end, it is imperative that the devices should be ,is section introduces a communication environment, a
manufactured with the ability to en/decryption functions. designated trust border, and outline the purpose of our
,is will ensure that malicious entities cannot physically suggested e-health systems that utilize IoT-cloud technol-
retrieve any raw data from IoT device directly. ,ere are ogy. ,en, we illustrate the detailed communication steps for
multitudes of security, privacy, and safety risks associated the proposed system that contains both discreet initializa-
with IoT-device utilization in e-health systems. One example tion and authentication phases.
is that personal data can be stolen and used to harm users or As mentioned in the literature review, e-health systems
patients in different ways. Another example is that bad that utilize IoT-cloud technology have the capability of
actors can initiate an insider/outsider attack to acquire managing a large variety of data, such as heart rate, tem-
patient registration information and biometric data. In this perature, diagnosis, and outcomes. ,is data can be analyzed
paper, the user authentication process is incorporated into to identify and calculate the severity of particular diseases.
cloud computing to protect against spoofing attacks. e-health systems leverage the capabilities of various hard-
ware devices and software components to achieve the or-
2.2.4. Resistance to Man-in-the-Middle Attacks. One of the ganized, controlled, and universal cloud computing
primary security vulnerabilities to overcome is man-in-the- incorporation. ,is enables fast/accurate transfer and data
middle attack resistance during authentication. An immoral analysis. Moreover, it ensures high overall system quality.
actor could possibly intercept transmitted authentication ,e analyzed data could be utilized to create graphic vi-
messages to spoof the communicating devices into believing sualizations and warnings corresponding to a given patients
they are legitimate users. ,is allows the attacker to com- health condition and stored confidentially. Moreover, data
municate with the server as an authorized user and collect or integrity and authentication are ensured through the utili-
modify data. Spoofing can also be used against real au- zation of cloud networks which implement appropriate
thorized users. For example, the attacker can mimic a le- security mechanisms for a given system.
gitimate server and communicate with the authorized user
to steal their credentials. Embedding all communicating
3.1. -e Communication Environment. Here, we introduce
device IDs into the protocol messages for device authenti-
our proposed communication environment for e-health
cation is an effective method of preventing man-in-the-
systems that utilize IoT-cloud technology. ,ere are three
middle infiltrations. Yu et al. [18] introduced an authenti-
essential components of our e-health communication:
cation/key consensus protocol in an IoT-cloud-based set-
wearable body biosensors, LPU (Local Processing Unit), and
ting. ,ey designed registration/authentication stages that
BSN (Body Sensor Networks) server. Wearable body bio-
utilizes a fuzzy-verifier technique as well as added verifi-
sensors are always fixed onto users and classified as edge
cation within cloud servers. ,is method can effectively
devices. ,ese edge devices collect and transmit biodata
resist insider attacks and DoS assaults.
from a given patient to LPU and BSN server for analysis. ,is
communication is highly efficient and enables fast and ac-
2.2.5. Multiple Security and Privacy Properties Must Be curate diagnosis and treatment in which real-time biodata
Managed Simultaneously. ,e security/privacy of both data such as electrocardiograph (ECG), electroencephalograph
and devices are the primary concerns of any e-health systems (EEG), electromyograph (EMG), and blood pressure (BP)
that utilize IoT-cloud technology. ,is is due to the wireless are retrieved. After the data is collected, the BSN servers can
communication protocols utilized by most BSNs, which tend process the data and provide bespoke treatment recom-
to be inherently insecure. ,is insecurity leaves the system mendations for patients in a timely manner. ,is reduced
vulnerable to outside attacks that could cause grievous harm. delay in diagnosis, and the treatment would greatly improve
,e following are three of the most common principles for patient outcomes. Our suggested communication frame-
protecting IoT-cloud-based e-Health systems from outside work for e-health systems that utilize IoT-cloud technology
 2037, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1155/2022/5300253, Wiley Online Library on [24/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Security and Communication Networks 5

Table 1: Notations.
Symbol Definition
P, Q Principals
X, Y Statements
K Long-term secrets or secret keys
P believes X P believes that X is correct
P sees X A message is transmitted that contains X to P
P said X P transmits a message that includes X within the present protocol or beforehand
P controls X P holds authority over S; P exerts control over X and should be believed
Fresh (X) X has not been transmitted in a message prior to the present protocol session
K
P↔ Q Key K is shared among both P and Q
{X}K Represents formula X, encrypted/secured under key K
BSi Identity of wearable biosensor i
LPUj Identity of the local processing unit j
Server Identity of the BSN server (cloud computing)
IDi Public identity of BSi
IDj Public identity of LPUj
IDs Public identity of server
IDdj Private identity of LPUj
IDss Private identity of Server
AIDi One-time alias identity of the wearable biosensor i
SID A set of unlinkable shadow identities SID � 􏼈sid1 , sid2 , . . . , sidn 􏼉
Trseq Track sequence number
Ris, Ri, Rj, Rsid1, Rsidn Random number
⊕ Bitwise operators
‖ Concatenation operation

employs a designated LPU and numerous biosensors to achieve expedient identification of BSi and avert replay attacks,
complete advanced registration on a designated BSN server. a track sequence number Trseq is designated. During each
Once advanced registration is completed, assigned security authentication session, Trseq must be altered and the new Trseq
authorizations shall be distributed and held onboard the will be stored both on the BSi and Server. During the au-
biosensors, designated LPU, and designated BSN server. thentication session, the freshness of inbound BSi requests can
Moreover, assigned security authorizations are utilized to be verified and directly identified by the Server through the use
establish protected channels of communication channels of Trseq stored in a designated backend database. ,en, the
and to authenticate devices. ,is will ensure that data Server automatically rejects any inbound demands and then
confidentiality and data integrity are maintained. terminates communication if the integrity of the Trseq in the
Our proposed system is comprised of two phases: a demand is not preserved in the backend database. If the de-
system initialization phase and an authentication phase. In mand remains intact, the Server instructs BSi to transmit a new
the initialization phase, approval will be determined for all request while simultaneously embedding a fresh SID as the
required security elements then distributed between the anonymized identity of BSi . Lastly, the Server issues a security
communication objects (body sensors, LPU, and BSN certificate (IDi , Kis , SID, Trseq ) to BSi . Meanwhile, the Server
server) through protected channels. After that, the au- maintains the same security certificate (IDi , Kis , SID, Trseq )
thentication phase will be utilized for the protection of all corresponding to each BSi in the backend database. Moreover,
exchanged communication/data between communicating registration between the Server and LPUj is accomplished
devices. Before outlining our proposed communication through a similar protocol; LPUj transmits its identity IDj in
protocols, we comprehensively illustrate the symbols and the form of a registration request to a designated Server. ,e
abbreviations utilized within this paper (Table 1). designated Server then calculates Kis � (IDs ‖Ris ‖IDi ) uti-
lizing a freshly generated public key Rjs . It then shares a se-
curity certificate (IDs , Kjs , IDj ). Lastly, the Server maintains
3.2. System Initialization. Firstly, body biosensor BSi trans- the security certificate (IDs , Kjs , IDj ) and its corresponding
mits its device identity IDi as a registration request to a LPUj in a backed database.
designated BSN server (cloud-based server). Once the request is ,e compiler will verify the token and either deny au-
received from BSi , the designated Server will then generate a thorization or add the sensor to a list of authorized devices. ,e
random number (Ris ) then utilizes its unique identity IDs to compiler is connected to the physical LAN network or wireless
calculate the secret key, Kis � (IDs ‖Ris ‖IDi ). Subsequently, network to receive the token key, save it to the device list, and
the designated Server will compute multiple unlinkable transmit the encrypted token key and data to authorized
shadow identities SID � 􏼈sid1 , sid2 , . . . , sidn 􏼉 for BSi . ,en, to network nodes. Once the token key is verified and validated,
 2037, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1155/2022/5300253, Wiley Online Library on [24/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
6 Security and Communication Networks

Cloud

Sensor Node Web Setting API Data Base Web Dashboard

Manual Mapping Get Mapping New User/Admin

Node Id + identify Device list + Information +
Card Number + Identify Card + Device List
1.New Sync Get Token Key User Id User Id Select User/
1.1.Push Sync Save To Device List Select Data admin
With 2M omni Area Get Information Information Information
Broadcast Token
Key User id = identify Insert Into New
Send Node Encrypt Card Number, User Id, Node id,
1.2.Save Node Token Key initial Information Sensor id, Time,
Token Key initial Values=0,
Save New (User Id Status
+ Node + Sensor)
Save to Database Select Data with
2.Sync again Check Sensor in list User Id, Time,
Power On Witch 2M Get Analyst Data Select Analyst Values, Status
omni Area Auto Data
Sync Node
2.1 Get Data Sensor Decrypt Data Get Data / Insert into Data
Encrypt Data witch Decrypt Node, Sensor,
twin Token Node + Node>Sensor Time, Values,
Sensor Save Transection Status

Figure 1: Authentication phase.

the token key is reset to generate a new record. Individual body Stage 2. LPUj ⟶ Server: MA2 � 􏽮M2 , Rj , IDj , V1 , MA1 􏽯.
sensors will receive authorization from the compiler and start Once the authentication request from BSi is received,
gathering data. Gathered data will be encrypted and trans- LPUj will produce a random number Rj and calculate M2 �
mitted wirelessly to the compiler and stored. (Kjs ⊕Rj ) and V1 � (MA2 ‖Trseq ‖IDj ‖Rj ‖Kjs ). ,en, LPUj
sends MA2 � 􏽮M2 , Rj , IDj , V1 , MA1 􏽯 to the Server.
3.3. Authentication Phase. In our proposed e-health systems
that utilize IoT-cloud technology, we assume healthcare Stage 3. Server ⟶ LPUj : MA3 � 􏼈Rs2 , Tr, V3 , V2 􏼉.
professionals with intelligent devices, endeavor to deliver Once the Server obtains MA2 � 􏽮M2 , Rj , IDj , V1 , MA1 􏽯,
instantaneous healthcare benefits through the use of an then the Server initially confirms if the tracking sequence
autonomous, noncontact information gathering and re- number Trseq remains present within a given request.
covery apparatus. Given that our communications network Supposing Trseq has concluded in MA2 , the Server executes
based on IoT is public, an authentication protocol that is condition (1), or else condition (2) shall be invoked.
both strong and resilient is necessary to protect information
exchanged between the biosensors, LPU, and BSN server. (i) Condition (1): test Trseq legitimacy; then seek an
Due to the public nature of hospital networks, the network is analogous tuple through Trseq from a designated
inherently insecure. Accordingly, a robust IoT communi- backed database. If Trseq legitimacy is confirmed,
cation architecture is required for the creation of a secure the Server recovers Kis . If legitimacy is rejected, the
channel that can be utilized for the safe exchange of data Server will terminate the current session. Alterna-
between BSi , LPUj , and the Server. tively, the Server shall validate M2 , V1 , and AIDi
,e authentication phase communication protocols are utilizing these subsequent algorithms:
detailed in Figure 1. (ii) Is the acquired M2 equivalent to calculate M2
expressed as M2 � (Kjs ⊕Rj )?
Stage 1. BSi ⟶ LPUj : MA1 � 􏽮AIDi , (M1 , Ri , Trseq , IDj )􏽯
BSi first generates random number Ri and calculates (iii) Is the acquired V1 equivalent
�� to calculate MA1
M1 � (Kis ⊕Ri ) and AIDi � (M1 ‖IDj ‖Trseq ). Next, BSi expressed as MA1 � (Trseq ���IDj ‖Rj ‖Kjs )?
sends MA1 � 􏽮AIDi , (M1 , Ri , Trseq , IDj )􏽯 as an authenti- (iv) Condition (2): in the event that the Server is unable
cation request to LPUj . If the shared Trseq value between the to locate Trseq within the request (MA2 , MA1 ), the
Server and BSi is not synchronized, BSi selects a new sid1 Server shall determine the validity/freshness of
from the master SID list and consequently sets sid1 as AIDi . AIDi � sidi from SID. In the event that the Server is
Lastly, BSi transmits an authentication request unable to recognize sidi harvested from the backend
(MA1 � 􏽮AIDi , (M1 , Ri , IDj )􏽯) to LPUj . database, the Server eliminates all connections and
 2037, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1155/2022/5300253, Wiley Online Library on [24/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Security and Communication Networks 7

then directs BSi to attempt validation once more (iv) Statement 4: Suppose P observes (X, Y) then P
utilizing a different shadow identity sidi . observes X. In addition, provided that P believes
K
P↔ Q and P observes {X}K , then P observes X.
In the event that at least one inspection agrees, the Server
shall produce random numbers Rs1 , Rs2 and then designate (v) Statement 5: Supposing a given formula contains
Rs1 as the new tracking sequence number Trseq (Trseq−new one component that is fresh, it is assumed that
�� �
Rs1 ). Subsequently, the Server computes Tr � (K � the entirety of the formula is similarly fresh. Sup-
�� is �Rs2 )
⊕Trseq−new , V3 � (Tr��Kis ), V2 � (IDj ‖Rs2 ‖Kjs ), SKis � posing P believes fresh (X), then P believes fresh
�� (X, Y).
(Kis ‖Tr), and SKjs � (Kjs ���IDj ‖Rs2 ‖Rj ). It should be noted
that the designated session key is SKis and shall be applied to Prior to analyzing the authentication process of the BSN,
the subsequent protected commination between BSi and the the following constraints must be considered:
Server. Moreover, SKjs is designated as the session key that (i) Constraint 1: BSi and Server believe BSi
will be utilized by the LPUj and Server. ,ereafter, S IDi ,Kis ,SID,Trseq
transmits MA3 � 􏼈Rs2 , Tr, V3 , V2 􏼉 to the LPUj as a response. ↔ Server.
IDj ,Kjs
(ii) Constraint 2: LPUj and Server believe LPUj ↔
Stage 4. LPUj ⟶ BSi : MA4 � 􏼈Rs2 , Tr, V3 􏼉. Server.
After receiving MA4 � 􏼈Rs2 , Tr, V3 , V2 􏼉, LPUj first cal- (iii) Constraint 3: Server believes fresh (Ri , Rj ).
culates (IDj ‖Rj ‖Kjs ); next, it confirms that the accepted (iv) Constraint 4: BSi and LPUj believe fresh (Rs1 , Rs2 ).
value V2 and the calculated value (IDj ‖Rs2 ‖Kjs ) are
(v) Constraint 5: Server believes BSi controls Ri .
equivalent. In the event that reciprocal authentication be-
tween the LPUj and the Server is confirmed, bothLPUj and (vi) Constraint 6: Server believes LPUj controls Rj .
the Server approve the utilization of secure session key SKjs . (vii) Constraint 7: BSi and LPUj believe Server controls
Lastly, LPUj sends MA4 � 􏼈Rs2��, Tr, V3 􏼉 to BSi . Upon (Rs1 , Rs2 ).
obtaining MA4 , BSi calculates (Tr���Kjs ) and cross-references
,e tangible BSN authentication protocols and proce-
it with the obtained value V3 . In the event that� the two
� dures are defined below. Each symbol is defined, and the
values are equivalent, BSi obtains Trseq−new � (Kis ��Rs2 )⊕Tr
BSN communication protocol is illustrated in Figure 1.
and sets Trseq � Trseq−new ; the next new authentication
session will utilize this technique. Lastly, BSi calculates an (i) Stage 1: BSi ⟶ LPUj : MA1 � 􏽮AIDi , (M1 , Ri , Trseq , IDj )􏽯,
SKis � Kis ‖Tr, session key that is utilized in cooperation where M1 � (Kis ⊕Ri ), and AIDi � (M1 ‖IDj ‖Trseq ).
with the Server. ,is guarantees mutual authentication (ii) Stage 2: LPUj ⟶ Server: MA2 � 􏽮M2 , Rj , IDj , V1 , MA1 􏽯,
between the Server and BSi . where M2 � (Kjs ⊕Rj ), V1 � (MA2 ‖Trseq ‖IDj ‖Rj ‖Kjs ), and
MA1 � 􏽮AIDi , (M1 , Ri , Trseq , IDj )􏽯
4. Security Analysis (iii) Stage 3: Server ⟶ �� LPUj : MA3 � 􏼈Rs2 , Tr, V3��, V2 􏼉,
In this section, we present our security analysis in terms of the where Tr � (Kis ��Rs2 )⊕Trseq−new , V3 � (Tr��Kis ),
aforementioned system criteria. ,e analysis of authentication and V2 � (IDj ‖Rs2 ‖Kjs ).
is based on a logic model. Before we conduct the formal logic (iv) Stage 4: LPU�� j ⟶ BSi : MA4 � 􏼈Rs2 , Tr,��V3 􏼉, where
analysis on the proposed authentication mechanism, we define Tr � (Kis ��Rs2 )⊕Trseq−new and V3 � (Tr��Kis ).
basic constructs and logic postulates. Hereafter, P and Q range ,e official analysis of the BSN joint authentication
over principals, X and Y range over statements, and K ranges protocol is detailed below:
over long-term secret keys (i.e., Table 1).
(1) LPUj perceives MA3 � 􏼈Rs2 , Tr, V3 , V2 􏼉; based on
Stage 3, it is apparent that LPUj has received and
4.1. Claim 1: Robust Access Control Can Be Achieved through perceived 􏼈Rs2 , Tr, V3 , V2 􏼉.
Mutual Authentication among Communication Entities in a IDj ,Kjs
BSN. We demonstrate some basic logical assumptions and (2) LPUj believes LPUj ↔ Server; based on Con-
constraints, where variables P and Q range over principals, straint 2, LPUj perceives that it shares IDj and Kjs
X and Y range over statements, and K ranges over long-term with the designated server.
secret keys. (3) Based on (1) and (2), LPUj perceives that the Server
K
(i) Statement 1: Suppose P believes P↔ Q and P per- said 􏼈V2 􏼉; therefore, we can derive that LPUj
ceives {X}K ; next, we presume that P believes Q said perceives that the Server said 􏼈V2 􏼉 utilizing State-
X. ment 1.
(ii) Statement 2: Suppose P believes fresh (X) and P (4) LPUj perceives fresh (Rj ); as Rj which is dispensed
believes Q said X; next, we presume that P believes by LPUj ; LPUj has the ability to evaluate the
that Q believes X. freshness of Rj and perceives (Rj ) as fresh if it meets
(iii) Statement 3: Suppose P believes Q holds dominion the specified criteria.
over X and P believes that Q believes X; next, we (5) Based on (3) and (4), LPUj perceives that the Server
presume that P believes X. perceives 􏼈V2 􏼉; if it is confirmed that LPUj
 2037, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1155/2022/5300253, Wiley Online Library on [24/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
8 Security and Communication Networks

perceives that the Server perceives 􏼈V2 􏼉 based on be circulated between BSi and the Server and then shall
Rule 2, then the claim is supported. be analyzed by the Server. Nobody is capable of tracing
(6) LPUj perceives that the Server governs 􏼈Rs2 􏼉; based or identifying BSi with the exception of the Server.
on Constraint 7, LPUj perceives that the random ,irdly, based on the following
�� equation expressed
number Rs2 is directly governed by the Server. as M1 � (Kis ⊕Ri )⊕Tr � (��Kis Rs2 )⊕Trseq−new and V3 �
(Tr‖Kis ), it is clear that these three variables are ran-
(7) Based on (5), (6), and Statement 3, LPUj perceives
domized through the use of the random numbers Ri and
􏼈V2 􏼉; therefore, it can be confirmed that LPUj
Rs2 . ,e random numbers cannot be recycled between
believes 􏼈V2 􏼉.
sessions. Given the aforementioned arguments, we believe
(8) BSi perceives MA4 � 􏼈Rs2 , Tr, V3 􏼉; in Stage 4, BSi is that the anonymity and untraceability of BSi can be ensured
confirmed to perceive MA4 � 􏼈Rs2 , Tr, V3 􏼉. and maintained.
(9) Based on Constraint 1, BSi perceives BSi
IDj ,Kjs ,SID,Trseq
↔ Server; therefore, BSi perceives that 4.3. Claim 3: Resistance against Spoofing Attacks and Data
IDi , Kis , SID, and Trseq are shared with the Server. Confidentiality Can Be Ensured. In the aforementioned unit,
basic authentication/login mechanisms that do not utilize
(10) Based on (8) and (9), BSi perceives that the Server session key agreements are insufficient and cannot ensure
transmitted 􏼈Tr, V3 􏼉; therefore, we can determine security in any way. ,e primary element required to ensure
that BSi perceives that Server transmitted 􏼈Tr, V3 􏼉 security is the establishment of two session keys. ,e first
utilizing Statement 1. session key is approved by BSi and the Server, and the
(11) Based on Constraint 4, BSi perceives (Rs2 ) as fresh; second key is appointed by the LPUj and the Server.
BSi accepts the freshness of Rs2 . Within the�� proposed BSN, two unique �� session keys,
(12) Based on (10) and (11), BSi perceives that the Server (SKis � (Kis ��Tr) and SKjs � (Kjs ‖IDj ‖Rs2 ��Rj )), will ulti-
perceives 􏼈Tr, V3 􏼉; it is guaranteed that BSi per- mately be created and utilized to secure data transmissions
ceives that the Server perceives 􏼈Tr, V3 􏼉 because of between BSi , LPUj , and Server. Moreover, we assert that all
Statement 2. the secure communication channels between BSi , LPUj , and
(13) Based on Constraint 7, BSi perceives that the Server Server are provided by the proposed scheme. Additionally,
governs 􏼈Rs2 􏼉; BSi perceives that the random the Server selects two highly volatile secrets (Kis , Kjs ) in
number Rs2 is directly governed by the Server. order to secure all communication transpiring within the
authentication phase of the proposed scheme. Without
(14) Based on (12) and (13), BSi perceives 􏼈Tr, V3 􏼉; knowledge or perception of the two selected secrets, it is
therefore, we can determine that BSi perceives exceedingly hazardous for bad actors to disrupt the proposed
􏼈Tr, V3 􏼉. authentication schemes or salvage any meaningful data from
,e finalized results of our simulation are shown below: communicated ciphertexts. ,erefore, confidentiality of the
data is ensured and maintained.
(i) LPUj perceives that the Server perceives 􏼈V2 􏼉 based
on (5).
4.4. Claim 4: Resistance against Common Electronic Attacks
(ii) LPUj perceives 􏼈V2 􏼉 based on (7).
(Man-in-the-Middle, Location Spoofing, and Replay) Can Be
(iii) BSi perceives that Server perceives 􏼈Tr, V3 􏼉 based Ensured. Bad actors could intend to deceive/bypass au-
on (12). thorized communication objects (BSi , LPUj , and Server)
(iv) BSi perceives 􏼈Tr, V3 􏼉 based on (14). through the use of imitation messages. ,e expedient and
efficient identification of counterfeit messages and the
Assuming the designated server is reliable/trustworthy
elimination/mitigation of possible threats within an IoT-
and based on the results of (5), (7), (12), and (14), both BSi
based healthcare setting is growing increasingly imperative.
and LPUj can authenticate each other on the designated
Messages counterfeited by bad actors come in a variety of
Server.
novel forms. Moreover, these malicious tricks could be
initiated from within the IoT systems inherently heteroge-
4.2. Claim 2: BSi Anonymity and Untraceability Can Be neous network architectures. It follows that an inspection
Ensured. During the authentication phase, we adopt ran- scheme with the capability of identifying fake communi-
dom numbers, Ri , Rj , Rs2 , and utilize them to randomize any cations and then, subsequently, preventing harmful attacks
communicated messages, namely, AIDi , M1 , M2 , V1 , Trseq , would be essential. In the proposed scheme, it is virtually
Tr, V2 , and V3 , where some are involved with BSi . Firstly, in impossible for a bad actor to forge genuine messages such as
the BSN, Trseq is employed as a unique single-use token that AIDi , M1 , M2 , V1 , Tr, V2 , and V3 in the absence of the
enables the rapid identification of BSi due to the fact that knowledge of Kis and Kjs . ,erefore, it is exceedingly dif-
Trseq does not contain any data relevant to BSi . Moreover, ficult to launch forgery attacks. Additionally, the Server
Trseq is actively refreshed following every successfully selects two highly volatile secrets because their volatile
completed authentication session. Trseq does not reveal any characteristics are highly resistant to brute-force attacks.
information about BSi . Secondly, a secret identity AIDi Assuming the bad actor successfully harvests and transmits
(occasionally a single-use/valueless si d shall be utilized) will validated recycled messages used in an earlier authentication
 2037, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1155/2022/5300253, Wiley Online Library on [24/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Security and Communication Networks 9

session to a potential target. ,en, a robust verification 5. Implementation

protocol utilizes a unique single-use key for all discrete
sessions. A cursory inspection of the random number’s ,is section summarizes the implementation of our pro-
freshness will determine their verification and validity. totype such that we could validate the practicality and
Consequently, the proposed scheme has incorporated a high usefulness of our proposed e-health system that utilizes IoT-
degree of replay attack resistance. Moreover, our proposed cloud technology. Our communication process requires
scheme, respectively, incorporates details of BSi and LPUj coordinated interaction among all the prototype hardware
into M1 and MA1 . In regard to untraceability, BSi is secure and software components. ,is communication protocol
within M1 . Only the Server is capable of executing an must be customized based on the intelligent devices,
untraceable attack on BSi , given that Kis is unknown to other transmissions, interoperability, internal protocols, access
entities. In regard to the Server, we present a robust veri- control requirements, security encryption/decryption re-
fication protocol that is capable of identifying location- quirements, and privacy requirements of any given system.
spoofing attacks. Within our proposed scheme, we incor- We implement security protocols at the component level
porated a natural resistance to location-spoofing attacks. (Raspberry PI, biosensors) and then determine the equiv-
Lastly, we examine our schemes’ toughness and ability to resist alent computational toll for each component of the system.
man-in-the-middle attacks. Moreover, we investigate if a bad Table 2 illustrates the implementation environment, in
actor could intrude on the authentication session and attempt which we utilized a Raspberry PI as the designated local
to deceive authorized communication objects by imperson- server that is connected to a wireless router/access point and
ating legitimate by utilizing forged/scavenged messages. Our a smart object (body sensor). ,e essential security elements
proposed scheme incorporates discrete communication ob- are all programmed using Arduino Software (C and Java-
ject identities into all protocol messages; this enables mutual Script) and executed natively on the Raspberry PI. Lastly, all
authentication between authorized communication objects: data is encrypted at the Raspberry PI and transmitted to a
Kis � (ID�s ‖Ris ‖IDi ), Kis � (IDs ‖Ris ‖IDj ),� M1 � (Kis ⊕Ri ), cloud API.
� �
Tr � (Kis ��Rs2 )⊕Trseq−new′, V3 � (Tr��Kis ), and First, we construct and execute a private network on a
V2 � (IDj ‖Rs2 ‖Kjs ) to ensure that such an event does not laptop computer with an Intel i5-6200U CPU, Nvidia
occur. It is evident that each discrete communication object’s GeForce 930M GPU, and 12 GB of RAM running Windows
identity is contained in all communication within the au- 10. Second, we utilize a TP-Link Soft Access Point to connect
thentication session. Forged/scavenged messages that contain the laptop and Raspberry PI via a Direct (P2P) WiFi con-
illegal identities are unusable due to the irrevocability our nection. ,e Raspberry PI has an ESP 8266 microcontroller
scheme’s authentication protocols. Moreover, previously sent and 125 MB RAM, which is capable of wakeup and packet
messages cannot be modified or recycled for reuse due to our transmission in less than 2 milliseconds and standby power
robust authentication protocol that utilizes two large random consumption of less than 1.0 milliamperes. ,e Raspberry PI
numbers with incorporated auxiliary values. Messages can be microcontroller can be programmed in C language to act as a
encrypted by anyone via the public key cryptosystem but can hub. ,erefore, we utilize the Raspberry PI as an intelligent
only be decoded by someone who knows the designated LPU object that can process and manage communication
random numbers. ,erefore, resistance against man-in-the- amongst IoT objects (body sensors) and cloud servers. ,ird,
middle attacks is ensured. we designed a client application to be implemented on user
facing devices such as smartphones, laptop/desktop com-
puters, or any devices that can run cloud API’s. ,is cloud
4.5. Claim 5: Simultaneous Management of Security and API stores the data using MySQL to control a database that
Privacy Properties Can Be Achieved through the Use of authorized users can access at any time. ,erefore, our API
Nonrepudiation Transaction Mechanisms. Within an IoT platform is easily customizable, flexible, and scalable and can
environment, a protocol with the ability to ensure that a be combined with our Raspberry PI platform to perform
collection of mobile objects, such as gateways and sensors, authentication. ,is system is then tested for practicability,
can coexist within the same time and place is integral at efficiency, and feasibility. Lastly, to balance system efficiency,
foiling probable fraud events. Our scheme’s protocol re- scalability, and security robustness, a secure RSA algorithm
quires the identity information for both BSi and LPUj to be [15, 16] that utilizes a random number generator and a
incorporated into all authentication messages (MA1 and bitwise exclusive operation are integrated into the system. In
MA2 ) to verify all object identities and the location of the order to be effective, the communications system channel
ongoing transaction session. According to our proposed must be coupled to at least one terminal that has an encoding
scheme, an authorized timestamp is contributed by the device and to at least one terminal that has a decoding
Server, MA2 � 􏽮M2 , Rj , IDj , V1 , MA1 􏽯 which will be utilized device. However, during the implementation phase of the
as evidence in the ongoing transaction between BSi and system, the values IDi , IDs , IDj , Ris , Rjs , si d, Trseq , Ri , Rj ,
LPUj after the validity of MA2 is established. ,erefore, it is Rs1 , and Rs2 are all set to a 32-bit random number generator
evident that the proposed scheme contains robust non- operation to ensure security density. Bear in mind that
repudiation transaction characteristics. within our experiment, IoT object access, is stringently
 2037, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1155/2022/5300253, Wiley Online Library on [24/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
10 Security and Communication Networks

Table 2: Implementation environment.

Environment Details
32-bit microcontroller Node MCU ESP8266
Micro 26 MHz to 52 MHz, memory 125 MB instruction
Raspberry PI
4 MB storage, built-in flash, 10 GPIO-Pins, SPI, I2C
Integrated TCP/IP protocol stack
Arduino software
Software platform
C, JavaScript
Cloud API PHP, HTML, MySQL

(a) (b)

Figure 2: (a) Client encryption/decryption. (b) Server encryption/decryption.

governed in accordance with the designated privacy policy. initialized, after which the last stage is the encryption/de-
In the absence of user consent, access to IoT objects is cryption processes. ,e system contains text-number con-
prohibited. versions that were executed based on ASCII coding/
Our implementation integrates readily available hard- decoding with extended lists that include 255 different codes,
ware devices, microcontrollers, and communication hard- as shown in Figures 2(a) and 2(b). Furthermore, we illustrate
ware to transmit data to the cloud system. ,e demonstrated our communication process utilized in our proposed
computation cost required in terms of the cryptosystem e-health system that utilizes IoT-cloud technology in
using a programmable microcontroller is reduced compared Figure 3.
to conventional methods. Our prototype utilizes Arduino In the following, we simplify our proposed system by
software that can be programmed to communicate in a dividing it into two main storage types: cloud computing
reader/sender configuration via the I2C synchronous serial and local storage, as presented in Figure 4.
protocol. ,e information will be sent bit by bit until the
(i) Local storage: it refers to dedicated IoT devices with
receiving device executes the request and transmits data
on-site or local network memory to store applica-
back while the I2C protocol allows for each authorized
tions, services, and collected data. ,is design im-
device to have its own unique address and for both devices to
proves reliability, interoperability, processing, and
take turns communicating over a single line. ,e compu-
transmission during offline operation. Local storage
tation cost of the proposed cryptographic operation in which
is a link between the Raspberry PI module storage
the experimental conditions are the same as that of [16]. In
and host system storage that allows for streamlined
addition, our Raspberry PI microcontroller board is pro-
protocols and local data processing. Moreover, RSA
grammed using Arduino software. ,en, we can program
encryption/decryption algorithms are integrated
control of the physical systems via read/write or analog/
into the local system to secure it from external at-
digital signals by integrating C programming, JavaScript
tacks, specifically, from attacks against the com-
programming, and RSA encryption and decryption to work
munication between devices via routers, Wi-Fi
in conjunction with Arduino software to significantly im-
networks, LAN networks, or cellular networks.
prove efficiency and security. Cryptography is defined as a
mathematical means by which data can be secured between (ii) Cloud computing storage: cloud API web services
two parties, whereby the sender and receiver use the same are applications, services, and physical devices that
designated algorithm for protection. ,erefore, potential can store collected data. Cloud computing reduces
intruders can neither understand the content of the mes- the implementation complexity of the system.
sages nor alter them. Any party that knows the designated Moreover, the security of the system is ensured
key (receiver) has the capability of decrypting the received through the utilization of RSA en/decryption at the
ciphertext. ,e complete design is divided into stages that cloud security layer. ,is ensures that only autho-
start by utilizing a random number generator to produce two rized users, such as administrative users, developers,
large prime numbers. ,en, a primality-testing unit is and patients, can access or manipulate stored data.
 2037, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1155/2022/5300253, Wiley Online Library on [24/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Security and Communication Networks 11

data

io n API Cloud Server

ypt Database Server
cr
En n
RS
A tio
en tica
uth
kA
Ac User/
Password
/Token
Router/WiFi/
Internet

Web Server

Authen/
n
Dataframe ke
s To
in/
User/Login/ og
/L
Token in
m
Ad

Sensor
Controller
BSN

USER Officer
Figure 3: Communication process.

Cloud Computing

Application Cloud Delivery

Error Handling
Storage Security Privacy policies
(RSA, AES) Private Public

Encoding/Decoding Access Control

Service Categories
User Profiling
(Admin, End User)
Presentation Layer Security and Privacy
User Physician Hospital
Cloud Security
Layer Security and
Application and Service
Layer Privacy Layer

Local Storage

Security Techniques
(Encryption/Decryption) Web Dashboard/
Interoperability
Wi-Fi/GPRS/
Storage Security 2G/3G/4G
(AES, RSA)
Pre-processing

Communication Security Access Layer

Transmission (Gateway, Router)
(Transcoding)
Raspberry Pi 3
ings Layer Preprocessing Layer Local Security Layer
(Intelligent Devices)

Figure 4: Main storage types for IoT devices.

,e presentation layer enables users and patients to custom user interfaces responsible for collecting data from
access their medical information via a web server or network-connected sensors and equipment. When a user
authorized Internet session. session is completed, any updates or modifications to the
data or information are synced with the cloud servers.
e-Health systems can utilize the aforementioned user in-
,e implementation architecture includes IoT sensor terfaces to obtain and assess the health condition of its users
devices, equipment, and application servers. It allows these then counsel them accordingly. Furthermore, the user in-
interconnected devices to communicate while utilizing terfaces are connected to cloud servers through the use of
 2037, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1155/2022/5300253, Wiley Online Library on [24/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
12 Security and Communication Networks

Identified by ID
and capture

Initial temperature
measurement

Students/Lectures/ Staffs
BSN
<37.5 C >= 37.5 C

Quarantine immediately,
trace back patient
Check basic symptoms: Check basic symptoms:
movements and disinfect
respiratory distress, cough, respiratory distress, cough, symptoms Monitor high risk
affected areas, monitor
sneezing, sore throat, sneezing, sore throat, groups and
any people that have had
nasal discharge nasal discharge quarantined patients,
direct contact with the
patient for symptoms. track all symptoms and
report to relevant
No symptoms health authorities.
No symptoms symptoms

Normal group:
Sending data via Cloud Sending data via Cloud
Come back to report for
Server and update Server and update
inspection within 14-20
patient status patient status
days

Method: POST API Cloud Server Method: POST API Cloud Server
(Temp + Patient ID + Authorize Keys) (Temp + Patient ID + Authorize Keys)

Cloud Server

Diagram : Infectious Respiratory Viruses (COVID-19) Line, WhatsApp, email Message Alert
(BSN) (Temp + Patient ID + Authorize Keys)

Figure 5: Utilizing IoT-cloud-based e-health system for COVID-19 detection and tracking.

middleware that facilitates intercloud communication with for the collection of live data and to deliver quicker, su-
the cloud database. ,e middleware utilizes cloud API web perior-quality healthcare services to patients. Further patient
services to transfer secure information between the web- data mining and analysis can be utilized to build predictive
based interfaces and the cloud servers. Moreover, middle- models. It is obvious that the implemented system offers
ware facilitates communication with the required databases robust security density with a reasonable computational
based on user-generated requests. ,e cloud API is spe- cost. ,e implementation simplifies the system, from the
cifically designed to control authentication, authorization, capture of data to the storage/analysis of data. Our system
and data processing. Lastly, data flow is consistent utilizes embedded sensors and equipment sensors controlled
throughout the layers and robust cloud security mecha- by an LPU node. ,e LPU processes the health data, displays
nisms, such as RSA, will be implemented at every layer. it locally, and transmits it to the cloud servers. ,e desig-
We implemented our system for the purpose of IoT- nated physician will receive daily emails with notifications in
cloud-based e-health management during the worldwide the application or webserver, containing the updated health
COVID-19 pandemic. ,e implemented system reduces the data of their patients. ,e application enables both the health
complexity of storing information, increases communica- care personnel and the patients to access and monitor rel-
tion efficiency, and improves the performance and accuracy evant health data. Additionally, security/privacy protocols
of a healthcare system. Results were verified using COVID- including the authentication of users and services, encryp-
19 spread as a case study scenario. ,e proposed imple- tion of communication, accountability tracking, and data
mentation architecture utilizes various web services and anonymization are utilized to secure the system from at-
security services which are exclusively cloud-driven and are tacks. Finally, users who interacted with the e-health system
further implemented as middleware. We analyze the security that utilizes IoT-cloud technology are allowed to track
of communication procedures to identify weaknesses so that potential COVID-19 symptoms. In addition, users can re-
improvements can be implemented to achieve the principal ceive alarms and visualizations of the collected live data and
security requirements. Intelligent body sensors were issued react to developing situations, as shown in Figure 5.
to patients in the field. ,is enables nurses/physicians to In our implemented system, first, the system identifies a
more effectively monitor and administer care to their pa- user’s ID and collects initial baseline measurements of the
tients. Nurses/physicians can utilize their mobile gateways user’s temperature, heart rate, blood pressure, and so on.
 2037, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1155/2022/5300253, Wiley Online Library on [24/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Security and Communication Networks 13

,en, it uploads the data to a designated private network and public and symmetric key cryptographic algorithms may be
sends it to the cloud API database. We utilize wearable body designed and verified based on design sets, such as DES,
sensors to collect temperature data over time. If a user’s data AES, and many others.
is abnormal, that is, body temperature measures greater than
37.5°C, then the user will be classified as possibly having Data Availability
contracted COVID-19, based on observed symptoms. We
separate symptoms into two categories: the normal group No data were used to support this study.
(no symptoms during the 14-day reporting period) and the
abnormal group (displaying symptoms during the 14-day Conflicts of Interest
reporting period). In the event that symptoms are confirmed
in a user, all that user’s data will be transferred to designated ,e authors declare that they have no conflicts of interest
Cloud API servers and their health status will be updated. regarding the publication of this paper.
Once a user displays COVID-19 symptoms, we recommend
immediate quarantine and contact-tracing. Additionally, Acknowledgments
disinfection of affected areas, active monitoring of any
people who had direct contact with the patient, and possible ,is work was supported by the Ministry of Science and
quarantine of contacted individuals are required to control Technology, Taiwan, grant nos. MOST 109-2221-E-259-011-
the potential spread of the virus. All patient data will be MY2, MOST 110-2926-I-259-501, MOST 110-2629-E-259-
continually collected, transmitted, and updated on our cloud 001, MOST 110-2218-E-011-007-MBK, 111-2218-E-011-
servers. Lastly, at-risk groups, such as physicians who have 012-MBK, and 110-2634-F-A49-004.
direct contact with many possibly infected individuals,
should track all symptoms and report them to relevant References
health authorities in a timely manner. Many users, including [1] N. Yousefnezhad, A. Malhi, and K. Främling, “Security in
physicians, nurses, and other stakeholders can access col- product lifecycle of IoT devices: a survey,” Journal of Network
lected data from the public cloud and private cloud via and Computer Applications, vol. 171, p. 102779, 2020.
authorized web portals. [2] J. J. Hathaliya and S. Tanwar, “An exhaustive survey on se-
curity and privacy issues in Healthcare 4.0,” Computer
6. Conclusion and Future Work Communications, vol. 153, no. 1, pp. 311–335, 2020.
[3] G. Gardašević, K. Katzis, D. Bajić, and L. Berbakov, “Emerging
,e growth/development of IoT applications has been ex- wireless sensor networks and internet of ,ings technologies-
ceptionally fast. ,is growth has proven to be useful in a foundations of smart healthcare,” Sensors, vol. 20, no. 13,
multitude of industries. Moreover, there have been pro- p. 3619, 2020.
[4] C. Butpheng, K.-H. Yeh, and H. Xiong, “Security and privacy
gressive changes in the administration and operation models
in IoT-cloud-based e-health systems-A comprehensive re-
of medical organizations across the world. Numerous in- view,” Symmetry, vol. 12, no. 7, pp. 1191–1235, 2020.
telligent IoT objects are all interconnected in order to create [5] X. Wang and S. Cai, “Secure healthcare monitoring frame-
a widespread IoT-based network architecture. However, a work integrating NDN-based IoT with edge cloud,” Future
new network helps expose new security challenges. Our Generation Computer Systems, vol. 112, pp. 320–329, 2020.
research has established a secure e-health management [6] S. Ray, H. Kathuria, K. Chakravarty et al., “Seronegative
system utilizing cloud computing and IoT-based BSN panencephalitis complicated by viral encephalomyelitis in a
structures. We propose an authentication mechanism to case of Good’s syndrome - a neuropathological report,” In-
ensure that major security requirements are satisfied. Based ternational Journal of Neuroscience, vol. 99, pp. 1–6, 2020.
on our implementation, our proposed system can be ef- [7] M. A. D. Shewale and S. V. Sankpal, “IOT based smart and
secure health care system Analysis & data comparison,” In-
fectively executed on low-powered LPU testbeds, such as a
ternational Journal for Research in Applied Science and En-
Raspberry PI, and the system ensures greater security than gineering Technology, vol. 8, no. 1, pp. 394–398, 2020.
historically available alternatives. Moreover, we performed [8] S. Chenthara, K. Ahmed, H. Wang, and F. Whittaker, “Se-
a rigorous formal analysis to confirm the security, privacy, curity and privacy-preserving challenges of e-health solutions
efficiency, and robustness. ,e implementation results in cloud computing,” IEEE Access, vol. 7, pp. 74361–74382,
prove that our IoT-cloud-based e-health management 2019.
system is practicable. We used Arduino microcontroller [9] M. Koutli, N. ,eologou, A. Tryferidis et al., “Secure IoT
units to implement and test RSA encryption and decryption e-health applications using VICINITY framework and GDPR
keys. ,erefore, it is evident that the security system is guidelines,” in Proceedings of the 2019 15th International
considered beneficial for sensor networks, as it can be Conference on Distributed Computing in Sensor Systems
connected while near a sink, particularly if it supports (DCOSS), pp. 263–270, IEEE, Santorini, Greece, August 2019.
[10] A. K. Chattopadhyay, A. Nag, D. Ghosh, and K. Chanda, “A
rechargeable batteries with easily interchangeable cells.
secure framework for IoT-based healthcare system,” in Pro-
However, system efficiency could be further improved with ceedings of the International Ethical Hacking Conference 2018:
the future development of more efficient leaner codes, al- EHaCON 2018, October 2019.
gorithms, and architectures. In the future, this design could [11] I. Alihamidi, A. Ait Madi, and A. Addaim, “Proposed
be modified to utilize a larger bit size for encryption and architecture of e-health IOT,” in Proceedings of the 2019,
decryption by modifying the algorithm. Lastly, numerous -e international conference on Wireless Networks and
 2037, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1155/2022/5300253, Wiley Online Library on [24/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
14 Security and Communication Networks

Mobile Communications (WINCOM), pp. 1–7, Fez, Morocco,

November 2019.
[12] A. M. Rahmani, T. N. Gia, B. Negash et al., “Exploiting smart
e-Health gateways at the edge of healthcare Internet-of-
,ings: a fog computing approach,” Future Generation
Computer Systems, vol. 78, pp. 641–658, 2018.
[13] K.-H. Yeh, “A secure transaction scheme with certificateless
cryptographic primitives for IoT-based mobile payments,”
IEEE Systems Journal, vol. 12, no. 2, pp. 2027–2038, 2018.
[14] K.-H. Yeh, “A secure IoT-based healthcare system with body
sensor networks,” IEEE Access, vol. 4, pp. 10288–10299, 2016.
[15] L. Ding, Z. Wang, X. Wang, and D. Wu, “Security information
transmission algorithms for IoT based on cloud computing,”
Computer Communications, vol. 155, no. 1, pp. 32–39, 2020.
[16] Q. A. Al-Haija, M. A. Tarayrah, H. Al-Qadeeb, A. Al-Lwaimi,
and A. Al-Lwaimi, “A tiny RSA cryptosystem based on
Arduino microcontroller useful for small scale networks,”
Procedia Computer Science, vol. 34, pp. 639–646, 2014.
[17] K. Stepien, A. Poniszewska-Maranda, and W. Marańda,
“Securing connection and data transfer between devices and
IoT cloud service,” Integrating Research and Practice in
Software Engineering: Studies in Computation Intelligence,
vol. 851, pp. 83–96, 2020.
[18] Y. Yu, L. Hu, and J. Chu, “A secure authentication and key
agreement scheme for IoT-based cloud computing environ-
ment,” Symmetry, vol. 12, no. 1, pp. 150–215, 2020.