Content uploaded by Shashank Shetty
Author content
All content in this area was uploaded by Shashank Shetty on Aug 11, 2020
Content may be subject to copyright.
Chapter 15
Fog Computing Application for
Biometric-Based Secure Access
to Healthcare Data
Sreekantha Desai Karanam, Shashank Shetty, and Kurup U. G. Nithin
15.1 Introduction
Healthcare 4.0 industry standards promote a patient-centric healthcare service
delivery at his doorsteps. Fog computing paradigm leverages us to deploy the power
of cloud computing at the edge devices in IoT networks to leverage cost-effective
communication, storage, and computations. “Fog computing is a term created by
Cisco that refers to extending cloud computing to the edge of an enterprise’s
network.”
“Fog Computing also is known as Edge Computing or fogging facilitates the
operation of computing, storage, and networking services between end devices and
cloud computing data centers.” Edge computing also enables security, mobility,
privacy, network bandwidth, and low latency features in IoT networks. These
features are essential to design a real-time healthcare management system. A
biometric security system ensures secure and perfect access to a system compared
to alphanumeric-based passwords in the digital world. Biometric security can be
further strengthened by adopting multi-model authentication. In the IoT domain,
every edge device will have a unique network address assigned by the IPV6 system.
Only the authorized end-users can access and control these edge devices to access
IoT-enabled services protecting their privacy. These edge devices provide secure
access for authenticated users through a biometric mechanism by verifying their
digital credentials in real time. This secure access mechanism can also be made
context-aware to understand user-specific requirements. High-security systems can
S. Desai Karanam () · S. Shetty
Department of CSE, NMAM Institute of Technology, Nitte, Udiupi Dist., Karnataka, India
e-mail: sreekantha@nitte.edu.in;shashankshetty@nitte.edu.in
K. U. G. Nithin
Department of CSE, VCET, Puttur, Karnataka, India
© Springer Nature Switzerland AG 2021
S. Tanwar (ed.), Fog Computing for Healthcare 4.0 Environments,
Signals and Communication Technology,
https://doi.org/10.1007/978-3-030-46197-3_15
355
356 S. Desai Karanam et al.
implement a combination of fingerprint, face recognition, and iris scanning to assure
the highest safety and privacy for authentication of the end-users.
15.1.1 Evolution of Technological Shift from Healthcare 1.0
to Healthcare 4.0 Standards
The major focus of Healthcare 1.0 is to reduce the manual and paperwork to enhance
productivity. Healthcare 2.0 focused on data processing and sharing with organiza-
tions. Healthcare 3.0 introduced patient-oriented IT solutions, while healthcare 4.0
objective was to provide real-time tracking of patient conditions and provide on-
site medical assistance. Healthcare 1.0 has progressed from the computerization
of healthcare data processing to real-time, on-site patient healthcare diagnostics
and context-aware AI-based solutions. The healthcare data/information sharing is
initiated within the hospitals and extended to the cluster of healthcare providers
across the country. Healthcare 4.0 promotes healthcare data sharing across the globe
conforming to technical and statutory standards. The technologies for Healthcare
1.0 were information technology solutions for hospital administration, extended to
Electronic Data Interchange (EDI), cloud computing, big data, big data analytics,
fog computing, IoT, Electronic Medical Records (EMR), wearable devices, block
chain, and artificial intelligence technologies. Today the major challenges faced in
Healthcare 4.0 are interoperability, conforming technical standards and ensuring
privacy, security, and confidentiality of data as per prevailing statutory framework
[38].
15.1.2 Fog Computing for Enhancing Biometric Security
The edge computing helps us to solve the difficulties connected with safety and
confidentiality of biometric signatures by improving security and privacy of critical
patient information. The intrinsic properties of fog computing permit additional
advantages of computing features which are essential for ensuring the privacy and
security-sensitive data access by computing important data at the fog nodes and
transmitting the secure and encoded data to cloud after processing.
15.2 Review of Recent Related Literature
Researchers have carried out an extensive survey of research papers on healthcare
4.0, fog and cloud computing applications from Springier and Elsevier publications
since the year 2007. The findings are revealed in the following discussion. A
15 Fog Computing Application for Biometric-Based Secure Access... 357
healthcare 4.0 architecture having three processing layers based on fog computing
was presented [1]. This architecture is used to implement an analysis of patient-
centric healthcare data. Two case studies were presented to confirm the effectiveness
of this system. It is mandatory to assure the security and confidentiality of data on
the health of a patient by law. In healthcare domain, providing secure access to
information is the most critical aspect.
The biometric features are used for authentication of end-users of health data,
since these personal features will not be forgotten, hired, purchased and also are very
hard to duplicate [2]. Ensuring data security in pathology labs is most crucial in the
design of smart medical systems [3]. The generation of a binary string of varying
length of bits, corresponding to the range of patient heart beat rate to incorporate
quality and security features in Wireless Body Sensor Networks (WBSN) using
Random Binary Sequences (RBS) generation method is called adaptive computing.
A protocol based on the Squared Secure Biometric Authentication (SSBA) metric
is used in a cloud platform. This protocol protects the confidentiality of the
critical data and ensures a safe identity in the cloud computing infrastructure [4].
Governments are promoting unique biometrics-based identification for their citizens
for authentication purposes to provides passport services, driving license, voter
cards, and public and social applications.
The demand is to develop a robust, secure and cost-effective authorization system
to protect privacy using multi-model biometric signatures [5]. After the successful
implementation of One Time Passwords (OTP), the application of biometrics that
is cancelable is highly recommended instead of conventional biometrics to promote
secure and private data sharing. Today cancelable or revocable biometric methods
are employed as these are more accurate and increase the privacy of data [5,6].
Considering biometric signatures and stored specimen templates as public data in
the authentication process leads to a weak security level, since the server stores
the original template and can be tampered by malicious hackers. This biometrics
sketch can be encoded as a binary string of constant size and saved in a tamper-
proof smart card to ensure better authentication. The suggestions on implementation
of Broadband Remote Access (BRA) systems to ensure a two-factor authentication
system to store additional data in a tamper-proof smart card as a second factor were
made by the authors [7].
A simple security protocol for securing the privacy of the data having cost-
effective packet transmission with three stages was proposed [8]. In the first stage,
the attacks are detected to eradicate the intruder attacks such as Wormhole, Sybil,
and Sinkhole. In the second stage, data is classified and ranked using WPM based on
its sensitivity and sends the nascent data to the MNS. In the last stage, an enhanced
Elliptic Curve Cryptography (ECC) would provide secure authentication services
between the interconnected end-users.
This method achieves the Packet Delivery Ratio (PDR) of 97%. The hardware
implementation of biometric-based authentication is very secure and withstands
regular attacks. This model was tested in real time and is applicable in many
domains [9]. The evaluation of the effectiveness of security to counter the wolf
attack was computed using WAP (Wolf Attack Probability). The attacker in the wolf
358 S. Desai Karanam et al.
attack tries to differentiate a user without knowing the user’s biometric signature.
The WAP provides the least level of security for the biometric authentication system
[10].
Ensuring the identity, privacy, and anonymity in a transaction using unique
biometric signatures are critical challenges. This protocol assumes that a secure
sketch and biometric signature template are publicly available entities. An end-user
need not have to save their private information and record data at the user’s sensors
level [11]. Designing of various models for integrating biometrics with smart cards
like On-Card Biometric Comparison, Store-on-Card, System-on-Card, and Work-
sharing procedure is experimented. These models are still in the design phase. Each
of these models is designed with a specific objective having pros and cons.
The authors discussed e-passports and Electronic Spanish National ID Card [12].
The design of security systems is carried out using hardware and software co-design
and it is also cost-effective [13]. This system identification is designed considering
the mobility of smart interconnected devices in ad-hoc networks. Application of
pi-calculus and the ProVerif verification tools are used in the design of biometric
authentication using Chen, Pearson, and Vamvakas (CPV) 02 protocol [14]. The
results have demonstrated that this protocol is effective, secure, and also correct
[15].
The study is carried out on user’s behavioral patterns, keystroke dynamics, and
texting of SMS messages, which are used as inputs for a multi-mode biometric
method in cell phones. The experimental results have revealed that these multi-
mode profiles can uniquely identify a user with high accuracy. A user authentication
protocol was designed using IBE (Identity Based Encryption) mechanism which
provides high security and improves WSNs authentication [16]. The proposal on
sensor forge resilience: liveness sensors and age-dependent sensors were explored
to check the feasibility of designing duplicate sensors to prevent forgeries of original
sensors [17].
A random orthonormal biometric remote authentication model which is not
susceptible to counter advanced attack in an open network was designed [18].
This model applies the user’s biometric signatures along with other authentication
factors to attain higher security levels. A protocol that is not anonymous is to
be designed to implement the services of anonymous and un-linkable for various
types of intruders [19]. This protocol is efficient in processing, ensuring security
and provides extended services. The bio-keys captured from different subjects
are quite random and unique to secure the Internet of Medical Things (IoMT)
[20]. These bio-keys are applied in medical data encryption to decrease resource
requirements. The results showed that this mechanism reduces the economics of
healthcare services and ensures safe medical data transfer between end-users and
service providers. A new authentication scheme for a cloud server was proposed
[21]. Experimental results of a security analysis performed on this model reveal
that this proposed scheme performs computationally and economically better. The
symmetric cryptography protocol shares end-to-end secret data between the nodes
having restricted resources of any particular remote entity [22]. The evaluation of
this protocol showed that it is safe, secure and saves the cost of energy.
15 Fog Computing Application for Biometric-Based Secure Access... 359
The proposal to permit an authenticated end-user to change the passwords and
biometric signatures without consulting the authorized administrator was presented
[23]. This proposal also gives a revocation policy to terminate the misbehaving
nodes in a network. The analysis of security aspects of this policy is carried out
using Burrows Abadi Needham (BAN) logic and random oracle model using the
popular Real-Or-Random model. The results derived from the Automated Validation
of Internet Security Protocols and Applications (AVISPA) tool revealed that this
model is tolerant to man in middle attacks. The proposal on the “Anonymous
Privacy-Preserving scheme with Authentication (APPA)” to IoT fog enabled model
is device-oriented [24]. Authors realized a multiple layered device authentication
using certification by pseudonym and anonymity. This scheme enables independent
update of certificate by SDs and pseudonym ensuring the confidentiality of data
sensed.
Cloud-based biometric authentication system BAMHealthCloud was discussed
[25]. A dedicated component of this system takes care of the security aspect.
ALGO Health Security check was performed on this model and results were found
satisfactory. The development of a secure healthcare framework is compatible with
cloud and mobiles platforms [26]. This framework considers security at inter-sensor
communication and patient’s data levels. This system employs multi-mode bio-
metrics signatures for inter-sensor communication through public keys. Evaluation
of this system showed that it is a feasible solution for future cellular healthcare
systems. System architecture based on smart e-health gateways with distribution
to verify the end-user’s authentication is proposed [27]. This method decreases
the workload on the medical sensors ensuring good security. This architecture
depends on Datagram Transport Layer Security (DTLS) handshake protocol which
is certificate-based. Analysis of security of this architecture reveals that it is highly
secure and also tolerant to Denial of Service Attack (DoS) attacks. A low weight
protocol having many features for remote end-user authentication was proposed
[28]. This protocol facilitates end-users to register using a gateway node in an IoT
environment.
After registration users can interconnect to the required sensor nodes through
IoT devices to avail every service directly. This protocol has less weight since
it is unidirectional and perceptual hash mathematical and XOR functions. This
protocol is less intensive in computing and hence very much suitable for the
IoT environment, where the capacities of resources are limited. The authors have
analyzed the security features using AVISPA tool to discover that it can tolerate
various security breaches. This protocol mainly concentrates on providing features
such as privacy, security, and integrity of the applications deployed in the cloud
environment. These applications are typically implemented on virtual machines
enabling trusted computing pools [29].
The Authenticated Key Agreement Protocol (AKAP) scheme for providing
security to remote users ECC in mobile client and server environments was
presented [30]. The authors discussed the security features of a random oracle model
with Elliptic Curve Discrete Logarithm Problem (ECDLP) and Curve Decisional
360 S. Desai Karanam et al.
Diffie-Hellman Problem (CDHP). Analyzing security features showed that this
scheme is tolerant of security threats.
The performance also revealed that this scheme is computationally less intensive
and has less communication cost compared to other schemes. A secure protocol
having a key establishment mechanism with mutual authentication in IoT-enabled
WSNs with better performance was designed [31]. The cryptanalysis conducted
on this protocol using the BAN logic model revealed that it tolerates various
security threats. A biometric template model for generating a biometric certificate
for user authentication was proposed [32]. This mechanism resists the users from
generating or extracting the other user’s biometric digital key pairs legally. This
model enables only the authenticated users to generate the keys. A quantitative
analysis of security features of Context Aware Security by Hierarchical Multilevel
Architectures (CASHMA), a multi-model biometric authentication system using
ADversary VIew Security Evaluation (ADVISE) formalism was carried out [33].
Authors studied the human factors that become threats for the authentication of
biometric systems [34]. FaceFirst is a tool to generate a completely automatic, easy
to use, face recognition system. This tool sends a message whenever a captured
face sample matches a face template stored in a database. The tool works in low-
resolution environments and enables real-time operations [35].
15.2.1 Literature on Iris Recognition based Biometric Security
The colored circular section in the human eye is called iris and this can be seen with
the normal eye. Iris is comprised of muscles that modify the pupil’s size and also
controls the quantity of light coming into the eye. Quantity of melatonin pigment
contributes to various colors in the formation of iris of humans. The iris muscle
foldings covering the ring generate a structure giving a greater level of detail. The
creation process of muscle structure is stochastic and it will not follow any specific
rules to govern the formation of structure in a human’s eye. This muscle structure
once created remains permanent throughout the life of the person. Each person’s
iris is unique and has a distinct pattern for each eye. These properties are considered
for individual recognition. A high-quality digital camera can scan the details of
iris muscle structures. The iris recognition system uses near-infrared (NIR: 700–
900 nm) radiation to capture iris structure. The iris recognition software is installed
in a dedicated system to get efficiency and security purposes. A camera captures
the image of this structure of iris muscles and its quality is improved by the image
enhancement procedures. Every iris formation is unique even the two iris of a person
are not identical and there are variations in iris of twins also [36,37].
This improved image is processed by the recognition system to identify the
distinct features to create a biometric template. Matching the sample current iris data
with this stored iris template confirms the identity of the person under consideration.
Iris recognition offers minimum cost of implementation with high security and user-
friendliness. Iris recognition has been implemented by border control agencies of
15 Fog Computing Application for Biometric-Based Secure Access... 361
the United Arab Emirates at border security checkpoints. All the foreign travelers
with visitor visas have to undergo an iris recognition system for entry into UAE.
CANPASS Air program based on iris recognition is operational in several Canadian
airports.
Aadhaar, a citizen identification system from the government of India’s program
is the unique method, where the biometric signatures are extensively used for citizen
identification and linked to all public services. Iris identification of a person using
iris is very effective in many applications.
15.2.2 Literature on Retina Recognition based Biometric
Security
The neural cells constitute a tissue of a thin layer in the eye called the retina. The
retina is situated inside of the human eye. The system of blood vessels carrying
blood to this thin layered tissue is represented by a specific structure, which serves
as a unique identification of a person. The structure of blood vessels is considered
distinct for every person. A special device is needed to capture this structure.
The high cost of usage and highly invasive nature of retina recognition makes its
less popular personal identification method and is only applied in highly secure
implementations such as defense and war fields. The infrared light having low
energy is used to capture the retinal patterns. The blood capillaries pass infrared
light and other tissues reflect this light. This reflected light is sensed and the
image is formed by the retina recognition system. This image is processed to
create a retina template representing the person’s retina signature. This retina image
capturing process is called biometric enrollment. The person’s identification may
be proved by capturing a new retinal sample and comparing it against the enrolled
retina template. Many government agencies like NASA, CIA, FBI, etc. are using
retinal recognition for personal identification purposes. Ensuring the privacy of
data, precise authentication, and anonymity of end-users identity is very critical
in healthcare domain. A framework for providing data abstraction with anonymity
of end-user is proposed using a paradigm named anonymous credentials. This
framework is implemented using blind signatures [39]. The survey on methods of
processing and storage of data in fog environment was carried out. This survey
revealed the various challenges and complications in carrying out fog data analytics.
Authors have designed a prototype to manage various parameters like ease of access,
scaling, communications and collaboration among the nodes, and non-homogeneity.
The functioning of this prototype has been explained using two cases [40].
362 S. Desai Karanam et al.
15.2.3 Statutory Requirements for Protecting the Patient Data
15.2.3.1 International Statutory Requirements
“Health Insurance Portability and Accountability ACT (HIPAA)” Standards are
developed by US. Department of Health and Human Services (HHS) to assure
the security and privacy of data and securing specific health information. The
HIPAA privacy rules and national regulations are designed to protect patient data
and other personal health data privacy of health programs, healthcare providers
and healthcare clearing centers. These entities share medical data to offer online
healthcare facilities. The objective of these standards is to establish mandatory
precautions to secure patient medical health data privacy.
They define terms for usage, sharing, and disclosing of patient data after
taking patient approval. These national regulations enable the patient to exercise
their rights over their health data records. The HIPPA security rule enforces
proper administration, infrastructural and technological precautions to guarantee the
privacy, unity, and safety of digitally secured health data [38].
HIPAA standards compliant business companies which are consuming delicate
and secured health data should enforce these standards in infrastructure, networks,
safety, regulations and operations [41,42]. China has also enforced many laws
and regulations for protecting healthcare data. The written consent of the patient
is mandatory for collecting, using, and sharing the medical and personal data. The
recent Cybersecurity Law enacted on May 1, 2017 prohibits the people of China
from using digital technologies to breach the privacy of patients and collect personal
information unlawfully. European Union (EU) has enacted General Data Protection
Regulation (GDPR) from 25th May 2018. This act prohibits any company from
gathering and processing medical data from EU or non-EU residents. Japan has
established Protection of Personal Information (APPI) act from 30th May 2017 [41].
15.2.3.2 Statutory Requirements for Healthcare Data Security in India
Govt. of India, Ministry of Health and Family Welfare is working on Digital Infor-
mation Security in Healthcare Act (DISHA). DISHA provides security, privacy,
standardization, and confidentiality for electronic health data. The aim is to set
up a National Digital Health Authority to exchange information related to health.
National Health Policy leverages National Health Information Network to share
Aadhaar mapped health records electronically. At present, Indian data privacy laws
are not planned for protecting the medical data. The section 43A of the Information
Technology Act, 2000 enforces general reasonable security practices, procedures
for sharing personal data which is sensitive [43,44].
15 Fog Computing Application for Biometric-Based Secure Access... 363
15.2.4 Consolidated Summary of Review of Literature
Authors have carried out the comparative study of all surveyed research papers
and highlighted the various methods, results, and applications of biometric-based
authentication systems as shown in Table 15.1.
15.2.5 Findings from Review of Literature
Authors after a survey of recent literature have discovered that many diverse
methods and protocols are used for authentication using multi-model biometric
techniques. The implementation using software, hardware using IoT microprocessor
boards in real time and experimentation was discussed in very few instances
of multi-model biometric signatures concerning Healthcare 4.0 standards. This
research gap has motivated researchers to research this area of authentication in
the healthcare 4.0 domain.
Authors found that biometric security systems have been successfully imple-
mented in banks, passports, visa offices, and many organizations. The collaboration
with KSHEMA Medical Colleges of Nitte Deemed to be University was planned
for successful implementation of this prototype. The authors have carried out the
experiments in labs and involving their staff and students for biometric signatures.
15.3 Proposed User Authentication System
The main objective of this chapter is to design a patient unique identification
system for secure access to patient healthcare data. Authors aimed at developing
a hybrid security mechanism at the edge devices to protect healthcare data from
intruders. Online home-based healthcare services are provided by smart hospitals to
patients at their homes. Patients shall enroll in accessing online healthcare solutions
from hospitals. The patients can avail of healthcare monitoring and consultancy
services from their smart home. The health conditions of the patient are recorded
and monitored by these smart wearable medical healthcare devices.
The wearable medical devices will sense abnormal patient’s health conditions
and send mobile notifications to the patient caregiver’s mobiles and hospital
authorities instantly. The patient’s healthcare data is private and sensitive, so
furnishing security and ensuring the confidentiality of this healthcare data is very
critical. Patient’s data needs to be accessed only by authenticated doctors and
hospitals securely, to the extent permitted by law.
Any violations in securing the privacy of healthcare data lead to breaching
of statutory obligations. This paper proposed a biometric authentication model to
verify that only authenticated users are accessing the data.
364 S. Desai Karanam et al.
Table 15.1 Consolidated Summary of Review Papers
Ref. No Methodology Applied Findings and Results Areas of Application
1. Healthcare 4.0 architecture Doctors can make intelligent healthcare
decisions in case of emergency in real time
Patient-centric medical data analysis
systems
2. Biometric features and fast identity
standards
Physiological biometrics are more stable
than behavioral biometrics
Smart medical systems
3. Uniqueness and randomness RBSs, is
measured using metrics of hamming
distances
This method is threefold faster for heart rate This method has significance for real-time
and intelligent healthcare devices.
4. Homomorphic encryption scheme Protects the confidentiality of the critical
data and ensures the safe identity
The biometric authentication process in the
cloud environment
5. Multi-model biometric signatures Promotes secure and private data sharing Passport services, driving license, voter
cards, public and social applications
6. Cancelable or revocable biometric scheme It has a minimal equal error rate compared
with those of the state-of-the-art techniques.
Suitable for IoT environments
7. Biometrics sketch is encoded as a binary
string
Better authentication. Tamper-proof smart card
8. Designed a model to detect Sybil, sinkhole,
and wormhole attacks under different
conditions.
Data packets delivery ratio is 97% and
cost-effective packet transmission
Ambulatory care unit for healthcare
9. Hardware implementation of
biometric-based authentication
Very secure and withstands regular attacks Applicable in many security domains
10. Secure sketch and biometric signature
template
Finger vein pattern algorithm to detect wolf
attacks
Secure biometric authentication systems
15 Fog Computing Application for Biometric-Based Secure Access... 365
11. Specially designed architecture and security
prototype for biometric authentication
policies
An end-user neither registers nor stores any
private data at the client sensor
Identity privacy and transaction anonymity
applications
12. System-on-card, store-on-card, on-card
comparing biometric data and sharing
The biometric data and processes are
protected using security methods
e-passports and electronic Spanish national
ID card
13. Interconnected security architecture for the
system of smart devices
Cost-effective, viable secure system for
smart devices that is 35% quicker and 5%
increased load efficiency
A security framework for security systems
14. Pi-calculus and the ProVerif tool CPV02 biometric authentication protocol,
is effective, secure and are also correct
Online banking
15. Behavior-based biometric methods for SMS
texting actions and communications
Matching fusion will improve the
classification accuracy by 8%
Create a reliable biometric security system
16. IBE mechanism
WSNs authentication
This protocol for authenticating a user has
better security features
Applicable in high-security wire fewer
sensor networks
17. Liveness sensors, age-dependent sensors Accuracy depends on applied unforgeability
and sensing technologies
Forge-resilient secure biometric systems
18. Unsusceptible biometric-based remote
authentication model
Countering advanced attacks in open
network
Complex computations are decreased
without losing accuracy in biometric
systems
19. An attribute-based non-anonymous and
fully anonymous scheme
Performance comparison of this scheme
with other similar schemes revealed its
out-performance
Appropriate for usage in
resource-constrained devices
20. Secure it, ECG data from 40 healthy patients
are collected and public data set, i.e.,
physionet
The outcome of the study revealed that this
method reduces process time and power
consumption
Real-time healthcare applications
(continued)
366 S. Desai Karanam et al.
Table 15.2 (continued)
Ref. No Methodology Applied Findings and Results Areas of Application
21. Sensing of patient heartbeats through ECG
signals using wearable healthcare devices.
Design of biometric security frames for
devices with resource constraints
The outcome of the study revealed that this
method reduces process time and power
consumption
This biometric security design has business
importance and relevant to society real-life
healthcare domain
22 Symmetric cryptography protocol Safe, secure and saves the cost of energy. Scheme performs computationally and
economically better
23 Authenticating users using ECC in wireless
sensor networks in the healthcare domain
Facilitates better security
Revoking stolen or lost cards and effective
security code words and updating the
biometric data. Dynamically adding sensor
nodes
e-authentication protocol using wireless
sensor networks
24 Anonymous privacy-preserving scheme with
authentication (APPA)
Model is tolerant to replay and man in the
middle attacks
Confidentiality in sensed data can be ensured
25. Biometric authentication system
BAMHealthCloud
BAMHealthCloud provides 0.12 EER, 0.98
sensitivity, 0.95 specificity
BAMHealthCloud, a biometric
authentication system for educational
healthcare, defense, and banking sectors
26 Secure healthcare framework which is
compatible with cloud and mobiles.
Security of inter-sensor communication
This framework is feasible future cell
phone-based healthcare applications
A ubiquitous and cloud-based security
framework for wearable healthcare solutions
27 Multi-mode biometrics signatures in
distributed smart e-health gateways
Facilitates important security features,
measures using extremely efficient key
creation methods
A feasible solution for future cellular
healthcare systems
28 DTLS handshake certificate-based protocol Highly secure and also tolerant of DoS
attacks
Intruder detection systems
29 Perceptual hash mathematical functions and
XOR instructions
Less intensive in computing and hence very
much suitable for IoT environment
Virtual machines and enabling trusted
computing pools
15 Fog Computing Application for Biometric-Based Secure Access... 367
30 AKAP scheme for providing security to
remote users ECC in mobile client and
server environments
The scheme is tolerant of security threats,
computationally less intensive
Mobile client-server environments
31 Key establishment for mutual authentication
in IoT
Cryptanalysis reveals that this technique has
improved performance and also tolerates
many security attacks
Secure data sharing protocol cloud
computing applications
32 BAN logic model Tolerates various security threats. The fingerprint data is used in Aadhaar card
ID, criminals identification cards ID, and
access control cards
33 Digital key creation and extraction
technique which are biometric-based
One can’t lawfully create or pull out other
users’ digital biometric key pairs
Biometric authentication systems
34 A multi-mode biometric authentication
system CASHMA’s security assessment
The security features are assessed with 0.1
confidence interval and 99% confidence
level
Biometric authentication systems
35 FaceFirst is a tool for face recognition Support vector machine Face recognition systems
36, 37. Iris recognition system Iris recognition is noninvasive and cost of
operations is low, user’s attention and
consent are essential
CANPASS air program is operational in
several Canadian airports.
The United Arab Emirates at border security
checkpoints
368 S. Desai Karanam et al.
Fig. 15.1 Patient data types
Patient Text
Data
Patient
Image Data
Patient Video
Data
Patient Audio
Data
Patient, diagnostic and
interventions text data
Photos, X-ray, scanning,
MRI, ECG images data
Patient Doctors Interactions,
Speech Therapy
Patient operations recording
and Physio therapy Progress
Recording
Authorized doctors can access these patient wearable devices for accessing
recorded healthcare data. After carrying out remote diagnosis, the doctors can also
interact and advise patients in real time.
15.3.1 Proposed Methodology
Authors proposed a fog computing based biometric solution for authentication of
end-users to access patient’s healthcare data. This healthcare data is managed in
a database mounted on the smart home server. Only authenticated end-users can
access this data. The healthcare data is multimedia data, and the patient’s details
such as name, address, and contacts comprise text data. The x-rays, ECG, and scan
reports are the image data. The patient’s and doctors’ interactions and speech theory
progress of patients are audio data. The video data consists of a recording of the
operations, the progress of physiotherapy exercises, etc. The health data types and
examples are shown in Fig. 15.1.
15.3.2 Stakeholders of Healthcare Data
The patient healthcare data would be used by different stakeholders as shown in
Fig. 15.2 for various purposes.
• The patient keeps track of his/her healthcare data for personal information and
monitoring purposes.
• Doctors would like to access the patient data for diagnosis, intervention, and
progress tracking purposes.
15 Fog Computing Application for Biometric-Based Secure Access... 369
Fig. 15.2 Stakeholders of healthcare data
• Clinical or pathology labs would like to access the health data for analysis and
reporting to doctors and patients.
• Hospital administration also accesses the patient’s data for billing.
• Govt. authorities also access the patient’s healthcare data for planning and
reporting purposes.
• The insurance company also accesses the healthcare data for processing the
insurance claims.
15.3.3 Architecture of Proposed User Authentication System
The authors have designed a layered architecture for providing authentication to
access the patient healthcare data securely as shown in Fig. 15.3.
15.3.3.1 Patient Data Layer
The patient is residing in his smart home or lying on smart bed when she/he is
not feel well. This patient health condition data is the innermost core layer in the
architecture (Fig. 15.3), where data is captured by the wearable medical devices
attached to the patient’s body or smart bed.
370 S. Desai Karanam et al.
Fig. 15.3 Layered
architecture
15.3.3.2 Edge Device Layer
The various health data capturing wearable medical devices are attached to the
patient body. These devices monitor the patient body conditions and record health
data continuously. Wearable medical devices are configured to send notifications to
patients, caregiver’s mobiles, and hospitals where a patient has enrolled in case of a
medical emergency.
15.3.3.3 Patient Cloudlet Layer
The health data captured is stored securely either in the patient’s mobile that plays
the role of secure storage edge space or cloudlet. This proposed system can also be
configured in such a way that the healthcare data can also be stored in the cloudlet
space dedicated to the smart home server of the patient.
15.3.3.4 Authentication Layer
The patient healthcare data needs to be shared by different end-users for different
purposes. Ensuring that only authorized users will access patient data to the extent
required and permitted by law is most important. In this context, authentication of
healthcare data plays a very important role.
The authors have designed a multi-mode biometric authentication system proto-
type for protecting this data.
The user’s authentication is provided by capturing text based on username and
password, a biometric image of the fingerprint, face recognition, and iris recognition
depending on the data type and significance of data that needs to be accessed.
15 Fog Computing Application for Biometric-Based Secure Access... 371
15.3.3.5 End-User Layers
The end-users of data are discussed and shown in Fig. 15.2 who can access
the healthcare data of the patient by proving their identity. The authentication
requirements for accessing the data varies on the criticality and type of data. The
text data is least critical and hence less secure, while image data and video data
have increasing levels of criticality and security.
15.3.4 The Block Diagram for End-User Authentication
The biometric authentication system block diagram is organized and nested blocks
as shown in Fig. 15.4.
15.3.4.1 Patient Smart Home Block
The innermost block is the patient smart home block which has four sub-blocks.
The patient sub-block comprises a patient with wearable medical devices. The edge
devices sub-block is to pre-process the captured data. The authentication sub-block
is to verify the identity of users and authenticate the user’s access.
Fig. 15.4 Block diagram of an authentication
372 S. Desai Karanam et al.
15.3.4.2 Hospital Online Services Block
The patients and old age people of smart homes have to enroll in accessing
online services with smart hospitals. The enrolled patients are provided with online
healthcare and consultancy services. The patients are connected to the hospital
through their medical wearable devices. Only authenticated doctors can monitor
and diagnose the patients online. The access to patient’s data and interactions with
health professionals is secured by the proper authentication mechanism. Hospitals
have to identify themselves with user code to sign in to patient data account. The
authorized doctors and medical professionals can retrieve the patient data to the
extent they are allowed to access after verification of their authentication.
15.3.4.3 Healthcare Services Providers
The insurance authorities, consultants, and Govt. authorities would also require
to access the patient healthcare data. These bodies can access patient data after
providing proper authentication online.
15.3.5 Safety and Privacy Levels of Data
The privacy and security of health data can be ensured by proper authentication
using biometric signatures. The authors proposed distinct levels of security for
various types of data as shown in Figs. 15.5,15.6, and 15.7. Text passwords and user
names represent the lowest level of security. Images and video recordings require
the highest level of security features.
Fig. 15.5 Security levels and
data types
15 Fog Computing Application for Biometric-Based Secure Access... 373
Fig. 15.6 Security levels, biometric signatures
Fig. 15.7 Integrated security levels, biometric signatures and end-users
15.3.6 End-Users Enrollment Procedure for Authentication
Purposes
15.3.6.1 Patient Enrollment
The hospital would broadcast the information on available online services through
their website. Patients shall register/enroll for accessing online healthcare services
on the website of the hospital. The patient’s fingerprint, face recognition, and retina
signatures are captured during patient registration.
374 S. Desai Karanam et al.
Fig. 15.8 Users biometric enrollment
15.3.6.2 Hospital Authentication
Hospitals will assign a unique identification code for patients who have enrolled
in online healthcare services. The hospital staff authentication codes and biometric
signatures are securely stored in patient smart spaces also. These authentication
codes are cross-checked and confirmed at the patient’s location before providing
access to the patient’s health data.
Fog computing has a very significant role in authenticating end-user identity to
ensure that authorized users are accessing the data. End-user’s biosignatures are
captured by fingerprint readers, face recognition and iris reader devices during the
enrollment process. These are images pre-processed and transformed into a standard
template with unique identification code and stored in the signature database in edge
devices after encryption and compression as shown in Fig. 15.8.
15.3.6.3 End-User Authentication Procedure
The end-users who would like to access patient data need to identify themselves
by providing their bioauthentication depending on the type and extent of data to
be accessed. These biosignatures captured from end-users are pre-processed. The
extracted features of the biosignature are matched with that of enrolled end-user
stored signatures. The system compares all the features of user specimen signature
with a stored template; if the exact match is found, then only the end-user is allowed
to proceed; otherwise, user’s data access request will be rejected. If authentication
is successful, the user will be able to proceed with data access. The entire process is
depicted in Fig. 15.9. The fog and cloud architecture for securing the patient data is
shown in Fig. 15.10.
15 Fog Computing Application for Biometric-Based Secure Access... 375
Fig. 15.9 Authentication verification process flow
Patient Cloudlet Fog
Computing User
Hospital
Smart Services
Private Cloud
Health
Insurance
Service
Networked
Hospitals Cloud
Smart Service
Providers
Healthcare Assisted Services
Fig. 15.10 Fog and cloud architecture diagram for securing patient data
15.3.6.4 Summary of Step-by-Step Procedure for Authentication
1. Hospitals publishing their smart healthcare online services
2. Patients subscribing for online smart healthcare services
3. Patient biometric enrollment process (finger, face, and iris signatures registra-
tion) for authentication purposes
4. Doctors and medical staff authentication for accessing the patient data
5. Patient registering their medical insurance company representatives and their
authentication
6. Hospital admin registration and authentication to access the patient data
376 S. Desai Karanam et al.
7. Govt. Public health representatives registration with hospitals for accessing the
patient data
8. Patient’s registration with networked smart hospitals to share patient healthcare
data
9. Registration of patient’s IoT smart home service providers for secure healthcare
data transmission
10. Registration and authentication of diagnostic labs for patient’s medical test data
15.4 Implementation of Proposed Methodology
The authors have designed an experimental prototype setup for verifying the
authentication of users using a face recognition technique shown in Fig. 15.11.
A web portal is designed to enroll users. Users use this portal for enrollment to
services as shown in Fig. 15.12. Raspberry Pi-3 and Pi camera are interfaced with
this web portal. The portal is developed in PHP and implemented Haar-cascade face
recognition for security purposes. The user’s face is exposed to Pi cameras during
the enrollment process. This system is trained to recognize the end-user’s face. The
face recognition algorithms extract the features of the end-user’s face and prepare a
specimen template.
During the testing phase when users are exposed to Pi camera the sample face
features are compared with stored sample face features. If the correct match is found,
then user face recognition is successful and the user is permitted to access the patient
health data, otherwise, the user is denied access to the data. The screenshots of
registered user recognition and display of user names are shown in Fig. 15.13.
The unregistered users not recognized are shown in Fig. 15.14. An alert message
and image of the unrecognized user are sent to the administrator as shown in
Fig. 15.15. To improve the accuracy of the recognition, maximum three testing
attempts are provided for the users for authentication. If more than three attempts
are made, then the notification is sent to system administration along with the
sample of the captured image and other details. Researchers are working on the user
Fig. 15.11 The experimental setup for face recognition
15 Fog Computing Application for Biometric-Based Secure Access... 377
Fig. 15.12 User registration into portal
Fig. 15.13 User recognition
authentication process using fingerprint and iris signature recognition on similar
lines of face recognition.
15.4.1 Conducting Experiments and Result Discussion
This authentication prototype is tested using the staff and students in our department
as users. The authors used a Raspberry Pi camera for capturing the face images of
end-users. Authors have selected end-users with different age groups, gender, and
categories. Ninety end-user’s biometric signatures were captured in this training
process. For each end-user, five face sample images with slightly different postures
are captured and stored in the edge device. At the testing phase randomly users
378 S. Desai Karanam et al.
Fig. 15.14 Unknown user recognition
Fig. 15.15 Alert to system administer
Table 15.2 Data set samples and results recorded
Persons Type Female Age Male Age Accuracy in %
Lab instructors 825–30 15 30–35 95
Faculties 12 28–35 15 28–40 96
Students 20 19–21 20 19–21 97
are called for face recognition. This system could recognize their faces accurately.
The face recognition accuracy achieved is about 95% and above. Authors are
experimenting with fingerprint and iris recognition modules and would like to
integrate all biometric signatures into one system. The details of the sample data
set and results in correct recognition are shown in Table 15.2.
15 Fog Computing Application for Biometric-Based Secure Access... 379
Authors have experimented with a set of users in the department and found that
the system gives satisfactory recognition results of about 95% accuracy.
15.4.2 Case Study and Challenges Faced in Fog Computing
Implementations in Healthcare 4.0
The patient is residing in his smart home or lying in the smart bed. The wearable
medical devices attached to his/her body will sense and transmit data about the
health conditions of the patient to edge device. The edge devices may be the smart
phone or any edge device with fog computing capability in the patient vicinity. This
edge device will process the data received from wearable devices and if any heath
data values are abnormal and critical then notifications will be sent to smart hospitals
authorities where patient has enrolled for smart services.
The doctors of smart hospitals can access the patient’s wearable devices data
after verifying their proper authentication. The edge will transmit the data to smart
hospitals. The hospital authorities will call the patient/caregiver and explain the
conditions of health and advice course of actions to be followed by the patient. This
patient heath data is also shared with clinical laboratories for further investigation
and accessing the patient data remotely with proper authentication.
Healthcare 4.0 facilitates remote monitoring of heath conditions of the patient
with human interventions. Patient can be altered and advised about medication by
the medical professional remotely through smart technologies.
The challenges in implementation are the following:
1. Optimizing the cost of healthcare
2. Infrastructure limitations to support real-time operations
3. Building trust in patients on smart healthcare services
4. Managing the interoperability of devices in fog and IoT network domains
5. Compliance with healthcare regulatory standards
6. Providing fool proof authentication mechanisms
7. Managing and sharing healthcare big data with high security and flexibility
15.5 Conclusions
Healthcare 4.0 standards leverage online health data sharing across the globe
conforming to technical and statutory standards. Healthcare 4.0 paradigm promotes
a patient-centric healthcare service delivery at his doorstep. The foolproof authenti-
cation mechanism is essential to prevent any intrusions into the healthcare systems.
Authors have carried out the comparative study of all surveyed research papers
and highlighted the various methods, results, and applications of biometric-based
authentications systems. A biometric security system is adopted to ensure secure
380 S. Desai Karanam et al.
access to a system. Biometric security can be further strengthened by adopting
multi-model authentication. This paper discussed national and international status of
healthcare data protection acts and tools used for biometric authentication. Authors
have discussed the prototype design for authentication of end-users of healthcare
data and carried out a face recognition experiment for authentication. The authors
have designed a layered architecture for providing authentication to access the
patient healthcare data securely. Authors have experimented with a set of users and
demonstrated satisfactory recognition results.
Future Scope Authors are collaborating with Nitte (Deemed to be) University
and KS Hegde Medical Academy, Mangalore, Karnataka for real-time patient data
management and data analytics. Authors are planning to implement this system in
the AB Shetty Dental College since authors have developed a web portal to capture
patient data which is being implemented in this hospital.
References
1. Kumari, A., Tanwar, S., Tyagi, S., & Kumar, N. (2018). Fog computing for healthcare 4.0
environment: Opportunities and challenges. Computers and Electrical Engineering, 72, 1–13.
https://doi.org/10.1016/j.compeleceng.2018.08.015.
2. Hamidi, H. (2019). An approach to developing smart health using the internet of things and
authentication based on biometric technology. Future Generation Computer Systems, 91, 434–
449. https://doi.org/10.1016/j.future.2018.09.024.
3. Wu, W., Pirbhulal, S., & Li, G. Adaptive computing-based biometric security for intelligent
medical applications. Neural Computing and Applications.https://doi.org/10.1007/s00521-
018-3855-9.
4. Kok Seng, Wong Myung, Ho Kim. (2012). Secure biometric based authentication for cloud
computing (pp. 86-101). Second international conference, CLOSER, Porto, Portugal, April
18-21. doi:10.1007/978-3-319-04519-1_6.
5. J. Wayman, A. Jain, D. Maltoni, D. Maio, An introduction to biometric authentication systems,
Biometric Systems. pp. 1-20, Springer, London, (2005), [Online]. doi:https://doi.org/10.1007/
1-84628-064-8_1.
6. Punithavathi, P., Geetha, S., Marimuthu, K., Hafizul Islam, S. K., Hassan, M. M., & Choo, K.-
K. R. (2019). A lightweight machine learning based authentication framework for smart IoT
devices. Information Sciences, 484, 255–268. https://doi.org/10.1016/j.ins.2019.01.073.
7. Sarier, N. D., Meadows, C., & Fernandez, G. C. (2012). Security notions of biometric remote
authentication revisited, STM, 2011. LNCS, 7170, 72–89. https://doi.org/10.1007/978-3-642-
29963-6_7.
8. Vaniprabha, A., & Poongodi, P. (2017). Augmented lightweight security scheme with access
control model for wireless medical sensor networks. Cluster Computing, 22(1), 1–12. https://
doi.org/10.1007/s10586-017-1669-7.
9. Maneesh, U., Anoop, M., Namboodiri, K., & Srinathan, C. V. J. (2009). Efficient biometric
verification in the encrypted domain, ICB 2009. LNCS, 5558, 899–908. https://doi.org/10.1007/
978-3-642-01793-3_91.
10. Une, M., Otsuka, A., & Imai, H. (2007). Wolf attack probability: A new security measure in
biometric authentication systems, ICB 2. LNCS., 4642, 396–406. https://doi.org/10.1007/978-
3-540-74549-5_42.
15 Fog Computing Application for Biometric-Based Secure Access... 381
11. Tang, Q., Bringer, J., Chabanne, H., & Pointcheval, D. (2008). A formal study of the privacy
concerns in biometric-based remote authentication schemes. In: L. Chen, Y. Mu, W. Susilo
(Eds.,) Information Security Practice and Experience. ISPEC 2008. Lecture Notes in Computer
Science (pp. 56-70), Berlin: Springer. doi: 10.1007/978-3-540-79104-1_5.
12. Sanchez-Reillo, R., Alonso-Moreno, R., & Liu-Jimenez, J. (2013). Smart cards to enhance
security and privacy in biometrics. In: Campisi P. (Ed.,) Security and privacy in biometrics
(pp. 239-274). London: Springer. doi: 10.1007/978-1-4471-5230-9_10
13. Awad, A. I., Hassanien, A. E., & Baba, K. (2013). A secure framework for OTA smart device
ecosystems using ECC encryption and biometrics. Berlin: Springer. https://doi.org/10.1007/
978-3-642-40597-6_18.
14. Salaiwarakul, M. D., & Ryan, C. L. (2008). Verification of integrity and secrecy properties
of a biometric authentication protocol. In W. Susilo (Ed.), ISPEC LNCS (pp. 1–13). Berlin:
Springer. https://doi.org/10.1007/978-3-540-79104-1_1.
15. Hataichanok, S., & Theoharidou, M. (2012). Multi-modal Behavioural Biometric Authen-
tication for Mobile Devices, SEC 2012. IFIP AICT, 376, 465–474. Retrieved from https://
link.springer.com/content/pdf/10.1007%2F978-3-642-30436-1_38.pdf.
16. Quan, Z., Chunming, T., Xianghan, Z., et al. (2015). A secure user authentication protocol
for sensor network in data capturing. J Cloud Comp, 4,6.https://doi.org/10.1186/s13677-015-
0030-z.
17. Phan, R. C. W., Whitley, J. N., & Parish, D. J. (2009). On the Design of Forgiving Biometric
Security Systems. In J. Camenisch & D. Kesdogan (Eds.), iNetSec 2009 – Open research
problems in network security. IFIP advances in information and communication technology
(Vol. 309). Berlin: Springer. https://doi.org/10.1007/978-3-642-05437-2_1.
18. Tran, N., & Dang, K. (2015). A multi-factor biometric-based remote authentication using fuzzy
commitment and non-invertible transformation. IFIP International Federation for Information
Processing, 9357, 77–88. https://doi.org/10.1007/978-3-319-24315-3_8.
19. Hamada, M., Ibrahim, S. K., Ashok, K. D., & Odelu, V. (2018). Attribute-based authentication
on the cloud for thin clients. Journal of Super-computing, 74, 5813–5845. https://doi.org/
10.1007/s11227-016-1948-8.
20. Pirbhulal, S., Oluwarotimi, W. S., Wu, W., Sangaiah, A. K., & Li, G. (2019). A joint
resource-aware and medical data security framework for wearable healthcare systems. Future
Generation Computer Systems, 95, 382–391. https://doi.org/10.1016/j.future.2019.01.008.
21. Jigna Hathaliya, J., Tanwar, S., Tyagi, S., & Kumar, N. Securing electronic healthcare records
in healthcare 4.0: A biometric based approach. doi: 10.1016/j.compeleceng.2019.04.017.
22. Abdmeziem, M. R., & Tandjaoui, D. (2015). An end-to-end secure key management protocol
for e-health applications. Computers, and Electrical Engineering, 44, 184–197. https://doi.org/
10.1016/j.ins.2019.01.073.
23. Challa, S., Das, A. K., Odelu, V., Kumar, N., Kumari, S., Khane, M. K., & Athanasios
Vasilakos, V. (2018). An efficient ECC based provably secure three-factor user authentication
and key agreement protocol for wireless healthcare sensor networks. Computers, and Electrical
Engineering, 69, 534–554. https://doi.org/10.1016/j.compeleceng.2017.08.003.
24. Guan, Z., Zhang, Y., Wu, L., Wu, J., Li, J., Yinglong, M., & Jingjing, H. (2019). APPA: An
anonymous and privacy-preserving data aggregation scheme for fog-enhanced IoT. Journal of
Network and Computer Applications, 125, 82–92. https://doi.org/10.1016/j.jnca.2018.09.019.
25. Kashish Shakil, A., Farhana Zareen, J., Alam, M., Jabin, S., & BAMHealthCloud. (2017). A
biometric authentication and data management system for healthcare data in Cloud, Journal
of King Saud University. Computer and Information Sciences, 32, 57. https://doi.org/10.1016/
j.jksuci.2017.07.001.
26. Farrukh Aslam Khana, Aftab Alia, Haider Abbasb, Nur Al, Hasan Haldar. A cloud-based
healthcare framework for security and patient’s data privacy using wireless body area net-
works.The2
nd International Workshop on Communications and Sensor Networks (ComSense-
2014). Retrieved from http://creativecommons.org/licenses/by-nc-nd/3.0/
382 S. Desai Karanam et al.
27. Sanaz Rahimi Moosavi, Tuan Nguyen Gia, Amir, Mohammad Rahmani, Ethiopia Nigussie,
Seppo Virtanen, Jouni Isoaho, Hannu Tenhunen. (2015). SEA: A secure and efficient authenti-
cation and authorization architecture for IoT based healthcare using smart gateways. Retrieved
from http://creativecommons.org/licenses/by-nc-nd/4.0/
28. Dhillon, P. K., & Kalra, S. (2017). A lightweight biometrics based remote user authentication
scheme for IoT services. Journal of Information Security and Applications, 34, 255–270.
Retrieved from https://daneshyari.com/article/preview/4955718.pdf.
29. Yeluri, R., & Castro-Leon, E. (2014). Identity Management and Control for Clouds. In Building
the Infrastructure for Cloud Security (pp. 141–159). Berkeley, CA: Apress. https://doi.org/
10.1007/978-1-4302-6146-9_7.
30. Mo, J., Hu, Z., & Lin, Y. (2018). Remote user authentication and key agreement for the mobile
client-server environments on elliptic curve cryptography. The Journal of Super-computing,
74, 5927–5943. https://doi.org/10.1007/s11227-018-2507-2.
31. Dheerendra, M., Vijayakumar, P., Venkatasamy, S., Ruhul, K., Hafizul, A., Islam, S. K., &
Gope, P. (2018). Efficient authentication protocol for secure multimedia communications in
IoT-enabled wireless sensor networks. Multimed Tools Applications, 77, 18295–18325. https:/
/doi.org/10.1007/s11042-017-5376-4.
32. Lin, J. L., Hsu, H. L., Jong, T. L., & Hsu, W. H. (2011). Biometric authentication. In P. S. P.
Wang (Ed.), Pattern recognition, machine intelligence and biometrics (pp. 607–631). Berlin:
Springer. https://doi.org/10.1007/978-3-642-22407-2_23.
33. Lee, H. W., & Kwon, T. (2007). Biometric digital key mechanisms for Telebiometric
authentication based on biometric certificate. In C. Stephanidis (Ed.), Universal Acess in
human computer interaction. Coping with diversity. UAHCI 2007 (Lecture notes in computer
science) (Vol. 4554). Berlin: Springer. https://doi.org/10.1007/978-3-540-73279-2_48.
34. Montecchi, L., Lollini, P., Bondavalli, A., & La Mattina, E. (2012). Quantitative security
evaluation of a multi-biometric authentication system. In F. Ortmeier & P. Daniel (Eds.),
Computer safety, reliability, and security. SAFECOMP 2012 (Lecture notes in computer
science) (Vol. 7613). Berlin: Springer. https://doi.org/10.1007/978-3-642-33675-1_19.
35. Michel Owayjan, Amer Dergham, Gerges Haber, Nidal Fakih, Ahmad Hamoush, Elie
Abdo. Face recognition security system, Springer, Berlin (2013). Retrieved from https://
www.researchgate.net/publication/259027363
36. Ali Alheeti, K. M. (2011). Biometric Iris recognition based on hybrid technique. International
Journal on Soft Computing (IJSC), 2(4). https://doi.org/10.5121/ijsc.2011.24011.
37. Shubhika Ranjan, Prabu S, Swarnalatha P, Magesh G, Ravee Sundararajan, Iris Recognition
System, International Research Journal of Engineering and Technology (IRJET), e-ISSN:
2395–0056, Vol: 04, Issue: 12, (2017). Retrieved from https://www.ijeat.org/wp-content/
uploads/papers/v8i5S3/E11030785S319.pdf
38. Chanchaichujit, J., Tan, A., Meng, F., Eaimkhong, S. Healthcare 4.0: Next generation
processes with the latest technologies. Retrieved from https://link.springer.com/book/10.1007/
978-981-13-8114-0
39. J. Vora, P. Dev Murari, S. Tanwar, S. Tyagi, N. Kumar and M. S. Obaidat. Blind signa-
tures based secured e-healthcare system International Conference On Computer, Information
and Telecommunication Systems (CITS), Colmar, 2018, pp. 1–5. Retrieved from https://
ieeexplore.ieee.org/document/8440186
40. Kumari, A., Tanwar, S., Tyagi, S., Kumar, N., Parizi, R., & Choo, R. (2018). Fog data analytics:
A taxonomy and process model. Journal of Network and Computer Applications, 128(2019),
90–104. https://doi.org/10.1016/j.jnca.2018.12.013.
41. Summary of the HIPAA Security Rule. HHS.gov. Retrieved November 30, 2019, from https://
www.hhs.gov/hipaa/for-professionals/security/laws-regulations
42. HIPAA Privacy Rule - HHS.gov. Retrieved November 30, 2019, from https://www.hhs.gov/
hipaa/for-professionals/privacy/index.html.
15 Fog Computing Application for Biometric-Based Secure Access... 383
43. The future of governance of health data in India. Ikigai Law. Retrieved November
30, 2019, from https://www.ikigailaw.com/disha-and-the-draft-personal-data-protection-bill-
2018-looking-at-the-future-of-governance-of-health-data-in-india.
44. DISHA and the draft Personal Data Protection Bill ... - Ikigai Law. Retrieved Novem-
ber 30, 2019, from https://marksmanhealthcare.com/indias-disha-different-global-patient-data-
protection-laws
