Volume 6, Issue 1, January – 2021 International Journal of Innovative Science and Research Technology

ISSN No:-2456-2165

Scrutinizing the Progress of Homomorphic

Encryption Scheme in Guaranteeing Data Privacy
Liz George Dr. Jubilant JKizhakkethottam
Dept. of Computer Applications, Dept. of Computer Science & Engineering,
St. Joseph’s College of Engineering & Technology, Saintgits College of Engineering & Technology,
Kottayam, India Kottayam, India

Abstract:- The privacy and security of data is always a II. WHAT IS HOMOMORPHIC ENCRYPTION?
major concern in the current era of big data and cloud
computing. Storing the data in encrypted form was the Homomorphic Encryption is an asymmetric encryption
approach followed to overcome this problem. technique with a pair of keys, where encryption is done using
Homomorphic Encryption evolved as a technique to a public key which can decrypted only using the
perform operations on encrypted data, providing the corresponding private key. It makes use of an algebraic
same results on working on raw data. It eliminates the system, allowing to perform functions on the encrypted data.
access of decrypted data by third party (cloud service The holder of the private key alone can access the
providers) for performing computation. Since its unencrypted data after the computation, allowing data to
inception in 1978, there occurred many contributions and remain secure and private while it is used by someone else.
enhancements in this area. The main breakthrough was One of the key features that distinguishes it from other
Fully Homomorphic encryption scheme proposed by encryption models is its resistance to quantum attack.
Craig Gentry in 2009. This paper discuss about the
classification of homomorphic Encryption Schemes as III. HOMOMORPHIC ENCRYPTION:
PHE,SWHE ,FHE and the additive and multiplicative CLASSIFICATION AND OPERATIONS
operations and about various schemes that have been
proposed from time to time in view of decreasing the There multiple types of encryption schemes associated
computational cost involved and to widen the range of with Homomorphic encryption, which performs diverse
applications/domains which implements HE. Various computations on cipher text. [2]
practical implementations of Homomorphic encryption
scheme are also discussed. 3.1 Classification
There are three variants of homomorphic encryption:
Keywords:- Homomorphic Encryption, Cloud Storage,
Privacy, Security. Partially homomorphic encryption allows performing
single operation either addition or multiplication ‘n’ number
I. INTRODUCTION of times on encrypted data

With the advent of numerous applications and Somewhat homomorphic encryption allows performing
automation of many traditional systems, there is remarkable different operations(addition and subtraction)but only limited
increase in the amount digital data available, which needs to number of times.
be stored without compromising the privacy of the data.
Many organizations are relying on Cloud Service Providers Fully homomorphic encryptioncombines the advantage of
for their data storage due to its affordable cost and low partially homomorphic encryption with somewhat
maintenance. The data is stored in cloud in encrypted form to homomorphic encryption, which allows to perform unlimited
retain privacy. To perform computations and analysis, CPS amount of operation for unlimitednumber of times
has to decrypt the data, resulting in the violation of the
organizations data privacy policy. 3.2 Operations
Homomorphic Encryption performs operations on
Privacy-enhancing Technologiesarea key research area encrypted data .The secret key for decryption is with the
in the current scenario. Homomorphic Encryption provides a client alone. After decryption, the result obtained is the same
promising solution in this regard [1]. It allows to perform as if the computation on Plain text.
computations on encrypted data, without compromising its
privacy. It can play a significant role in computation and
analysis of data in domains like healthcare and finance, where
the need to assure privacy is paramount.

IJISRT21JAN696 www.ijisrt.com 1413

 Volume 6, Issue 1, January – 2021 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
Additive operation achieving fully homomorphic encryption. It was a major
Encryption functionE is additively Homomorphic if breakthrough for researchers who were relentlessly working
for a solution to growing attacks on data privacy. He
E(m1 + m2) = E(m1) + E(m2) ∀m1,m2 ∈ M ,without describes it as a glovebox where everyone can get their hands
revealing m1 or m2. into the glovebox and handle what's inside, but nothing can
be taken out of the glovebox. The raw materials inside the
where M is the set of all messages glovebox can be accessed by all and they can build inside the
Multiplicative operation box. But only the person with key alone can retrieve the
Encryption function E is multiplicatively Homomorphic newly built product. But the main limitations of this work
Encryption if were the noise is attached with cipher text and increases in
each operation. In 2010, Smart and Vercauteren generated an
E(m1 ∗m2) = E(m1)∗E(m2) ∀m1,m2 ∈ M, without revealing encryption scheme with smaller ciphertext and key as an
m1 or m2 extension of Gentry’s original scheme [11] without
sacrificing the security. The main motivation behind this
where M is the set of all message scheme was conceptual simplicity.

IV. HISTORY OF HOMOMORPHIC ENCRYPTION The homomorphic cryptosystems which are popular
now is based on the scheme put forward by ZvikaBrakerski et
The phrase homomorphic is derived from Greek words al.[12].The encryption schemes introduced after this are
‘homos’, which means ‘the same’ and ‘morphe’ with known as second-generation FHE schemes, where growth in
meaning ‘shape’. The major limitation of systems using the noise during the homomorphic computation is slower.
encryption schemes are they can be used only for storing and
retrieving encrypted data. Performing computations on data The main highlight of homomorphic encryption is the
needs the decryption of data, which affects data security. The ultimate data privacy it offers, since computations can be
concept of Homomorphic Encryption was first proposed with performed within ciphertext, eliminating the burden of
the name “private Homomorphism’ in 1978 by Rivets, decryption. In the era of Internet, where, everything's public:
Ronald L., Len Ad leman, and Michael L. Dertouzos[3], after our browsing information, location, inclinations etc. the
the RSA encryption. But little progress was made on complete data privacy offered by homomorphic encryption
developing such a system for the next 30 years. has broad applications, especially in fields like Health care
and Finance.
4.1 Different Homomorphic Schemes
Several cryptosystems based private homomorphism V. APPLICATIONS
were proposed. But most of them were either additively or
multiplicatively homomorphic. A few which were fully The first and foremost application of Homomorphic
homomorphic, but had the limitation that ciphertext size encryption can be identified as the secure computation on
grows with the number of operations[4]. An additive data stored in cloud environment [13]. It ensures data privacy
Homomorphic encryption, which can encrypt only a single bit by allowing the cloud provider to perform computation on
was proposed by ShafiGoldwasser et al.[5] encrypted data.The IoT technology implementation in
healthcare has reformed personal health care by enabling
Additive-Multiplicative Homomorphism was remote monitoring of terminally ill patients using sensors
proposed[6], which guarantees that the result obtained after attached to them [14]. One of the important advantages of
performing computation on two encrypted values is cloud computing is the effective storage space it can provide
equivalent to the result found by encrypting the value for health records and the data provided by sensors , that can
obtained after performing computation on two unencrypted be shared with healthcare providers and researchers making
values. The main drawbacks were, absence of single system the medical research more effective. Meanwhile, using
which was additively and multiplicatively homomorphic and appropriate homomorphic encryption schemes, authorized
lack of functions which had compatibility with HE. entities can retrieve relevant information from the encrypted
health records eliminating the concern of data privacy. [15]
In 1999, Paul Paillier[7] proposed a partial fully
homomorphic encryption scheme which allows addition of Use of Genome sequences plays a significant role in the
two cipher texts and multiplication of a ciphertext by a study of complex diseases or epidemiology. But the broad
plaintext number. In 2005, a homomorphic scheme devised sharing of these data always raises threat as DNA sequences
by Boneh, Goh, and Nissim allowed both additions and are biometric identifiers, similar to finger prints. Any leakage
multiplications using a ciphertext with constant-size. of these information into wrong hands will have wide
However, only one multiplication is permitted, making the destructive impacts and it will be the utmost violation of the
system somewhat homomorphic[8]. identity privacy.Homomorphic encryption schemescan be
incorporated, ensuring the security of different genomic
The emergence of cloud computing technology in early datasets uploaded to the cloud and they can be effectively
2000 triggered the attention back on Homomorphic shared for studies to provide precision medicine and thus
Encryption[9]. In 2009 Craig Gentry’s thesis[10] on fully improving the health and wellbeing of patients [16].
homomorphic encryption, provided a basic outline for

IJISRT21JAN696 www.ijisrt.com 1414

Volume 6, Issue 1, January – 2021 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
By combining the properties of Blockchain technology REFERENCES
and Homomorphic encryption concepts, it is possible to
create an efficient e-Voting system which is public and [1] S. E. H. E. G. Maha Tebaa, "Homomorphic encryption
transparent, meanwhile protecting the identity of voters [17]. method applied to Cloud Computing," 2012 National
Using signature homomorphism scheme in Electronic voting Days of Network Security and Systems, 2012.
system can effectively manage the workload involved in
[2] C. B. C. C. G. A. J. Frederik Armknecht, "A Guide to
verification that occurs with the increase in the voters or
Fully Homomorphic Encryption," 2011.
contenders. [18] It also ensures privacy of voters and security
of voting. [3] L. A. M. L. D. Ronald L. Rivest, "On data banks and
privacy homomorphisms," in foundations of secure
Watermarks are used to show the identity of its owner, computation, 1978.
thereby ensuring the copyright.Finger printing is another [4] D. M. Freeman, "Homomorphic Encryption and the
scheme that prevents illegal copying of digital data. It helps BGN Cryptosystem," 2011.
to trace the source of an illegal redistribution, thus allowing [5] S. M. Shafi Goldwasser, "PrProbabilistic encryption and
copyright holder to take legal action.Homomorphic how to play mental poker keeping secret all partial
Encryption can be effectively used in digital watermarks and information," in proceedings of the14th Annual ACM
fingerprinting schemes, allowing to add a mark in to an Symposium on Theory of Computing,, 1982.
encrypted document [19].
[6] C. F. T. Tomas Sander, " Protecting Mobile Agents
Against Malicious Hosts," Mobile Agent Security, pp.
Homomorphic Encryption is capable of ensuring the
44-60, 1998.
security of sensitive data in financial sector. This may vary
from sensitive information pertaining to companies like [7] P. Paillier, "Public-key cryptosystems based on
Intellectual Property and trade secrets to the personal composite degree residuosity classes," Advances in
information of individuals. The capacity of homomorphic cryptology, pp. 223-238, 1999.
encryption to perform computation on encrypted data allows [8] E.-J. G. N. Dan Boneh, "Evaluating 2-DNF formulas on
to perform, secure search, analytics and collaboration of ciphertexts," in Theory of cryptography , 2005.
sensitive data and assets. [20] [9] L. X. W. B. Zhang Shaomin, "Study on the Protection
Method of Data Privacy Based on Cloud Storage,"
VI. CONCLUSION International Journal of Information and Computer
Science, pp. 46-51, 2012.
Homomorphic encryption had come great far ever since
[10] C. Gentry, "Fully Homomorphic Encryption Using Ideal
its launch in 1978. Different HE schemes have evolved based
Lattices," in Proceedings of the forty-first annual ACM
on application and security requirements. The biggest barrier
symposium, 2009.
that restricts the popularity of homomorphic encryption is its
lack of speed due to the large computational overhead [11] F. V. N.P. Smart, "Practice and Theory in Public Key
involved. The lattice-based cryptosystem implemented in Cryptography," 2010.
fully homomorphic encryption is sophisticated. Performing [12] Z. B. a. C. G. Vaikuntanathan, "Fully Homomorphic
even basic operations need massive computations and large Encryption without Bootstrapping," in International
cipher size.[21] Approaches for tuning the system to Conference on Information Technology Convergence
improve efficiency may result in compromising the security and Services, 2012.
of the system. The high cost involved in reducing the noise in [13] H. L. Maya Louk, "Homomorphic encryption in mobile
Fully Homomorphic Encryption Schemes is a major multi cloud computing," in International Conference on
apprehension. Even though, Fully Homomorphic encryption Information Networking, 2015.
schemes proposed can play a significant role in enhancing the
privacy of various applications, the efficient and cost [14] D. R. K. Anjali S. Yeole, "Use of Internet of Things
effective practical implementation of such systems in (IoT) in Healthcare: A Survey," in Proceedings of the
numerous platforms are yet to be evolved.Introduction of ACM Symposium on Women in Research , 2016.
FHE inculcated hope to solve other long waiting [15] T. S. Övünç Kocabaş, "Medical data analytics in the
problems(applications) such as Functional Encryption cloud using homomorphic encryption," E-Health and
(FE),Identity-based encryption (IBE) and Attribute-base d Telemedicine: Concepts, Methodologies, Tools, and
encryption[22]. Applications , 2016.
[16] A. G. Y. P. S. G. Marcelo Blatt, "Secure large-scale
genome-wide association studies using homomorphic
encryption," Proceedings of the National Academy of
Sciences of the United States of America, 2020.

IJISRT21JAN696 www.ijisrt.com 1415

Volume 6, Issue 1, January – 2021 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
[17] R. T.-M. C. M.-E. W. Jen-Ho Hsiao, "Decentralized E-
Voting Systems Based on the Blockchain Technology,"
in International Conference on Ubiquitous Information
Technologies and Applications , 2017.
[18] T. W. Q. Z. C. A. X. Xingyue Fan, "HSE-Voting: A
secure high-efficiency electronic voting scheme based
on homomorphic signcryption," Future Generation
Computer Systems, vol. 111, pp. 754-762, 2020.
[19] D. P. T. S. S. R. N. S. G. Dinesh Kumar, "An Efficient
Watermarking Technique for Biometric Images," in 7th
International Conference On Advances In Computing &
Communication, 2017.
[20] M. N. a. V. V. Kristin Lauter, "Can Homomorphic
Encryption be Practical," in Proceedings of the 3rd ACM
workshop on Cloud computing security workshop, 2011.
[21] L. Morris, "Analysis of Partially and Fully
Homomorphic Encryption," 2013.
[22] H. A. A. S. U. ABBAS ACAR, "A Survey on
Homomorphic Encryption Schemes: Theory and
Implementation".

IJISRT21JAN696 www.ijisrt.com 1416