0% found this document useful (0 votes)
15 views7 pages

E-Voting System with Homomorphic Encryption

This document discusses implementing an e-voting system using Paillier homomorphic encryption. It provides background on homomorphic encryption and discusses related work on using Paillier and other homomorphic encryption schemes for applications like e-voting and encrypted data processing. The paper then outlines the proposed e-voting model and methodology before discussing results and future work.

Uploaded by

ssabbir
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
15 views7 pages

E-Voting System with Homomorphic Encryption

This document discusses implementing an e-voting system using Paillier homomorphic encryption. It provides background on homomorphic encryption and discusses related work on using Paillier and other homomorphic encryption schemes for applications like e-voting and encrypted data processing. The paper then outlines the proposed e-voting model and methodology before discussing results and future work.

Uploaded by

ssabbir
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Proceedings of 4th International Conference on Computer and Management ICCM (2018)

E-Voting using
Homomorphic Encryption
Siddhant Bajaj 1, Vidit Chitkara 1, Abhishek Bindlish 1, Hardik Nangia 1, Dr. Rachna Jain 2
1
Research Scholar, Bharati Vidyapeeth College of Engineering, Delhi, India
2
Department of CSE, Bharati Vidyapeeth College of Engineering, Delhi, India
sbajaj1997@[Link], [Link]@gmail..com, [Link]@[Link],
hardik.nangia1997@[Link], [Link]@[Link]

Abstract- Cloud computing is regarded as the development and building of distributed computing,
virtualization technologies, grid computing and parallel computing but security and privacy issue a strong
barrier for users to adapt to cloud computing. Some of the most commonly faced issues with cloud
computing are Data Loss, Distributed-Denial-of-Service Attacks, Phishing, Shared Cloud Computing
Services and Inadequate Data Backups and Social Engineering Attacks and to curb such attacks and
prevent access to sensitive data, the data is often encrypted before storing it. The data needs to be decrypted
first before it can be accessed and therefore homomorphic encryption is adopted as it allows the cloud
provider to make operations on the data which is encrypted without decrypting it first. The encryption is
said to be partial homomorphic if it can perform a limited number of operations whereas it is called fully
homomorphic if it can perform both addition and multiplication. In this paper we implement and discuss an
e-voting system using paillier’s encryption.

We have implemented and analysed Paillier’s scheme, developing a secure and robust e-voting platform
which is remotely accessible.

Keywords: Homomorphic Encryption, Paillier Algorithm, E-voting


many advantages over various other computing
techniques and provides and improves the
1. INTRODUCTION
availability of IT services. Privacy and security
pose as the main hurdle to its fast adoption. Data
In this document, we will implement an e-voting
breach, hijacking of accounts, malware injections,
system that works on paillier encryption
insecure-APIs, denial of service attack and data
algorithm. Cloud is a space where computing has
loss are some of the many security and privacy
been pre-installed and exist as a service; data,
concerns related to cloud computing. Privacy
applications, operating systems, and storage exist
protection is a major technical challenge for
on the web. Cloud computing is a Pay-per-Use-
which various methods of encryptions have been
On-Demand mode for users that can conveniently
introduced. One such encryption technique is
access shared resources through the web. Where
homomorphic encryption where the system
the resources include network, server, storage,
performs operations on the encrypted data rather
application, service and they can be deployed
than decrypting it first and then performing
quickly and with ease. Cloud computing offers
computations. This results in fast computations
HTTPS://[Link]/LINK/[Link] ELSEVIER-SSRN (ISSN: 1556-5068) 96

Electronic copy available at: [Link]


[Type here]

and access to data, thus improving the efficiency 4. Verifiability: The votes should be
and maintaining the privacy of the users. The verifiable at any time and any place.
users or the clients are the only people who have In this e-voting system we’re going to use
access to the secret key. Broadly there are two Paillier’s algorithm to cast votes and securely
kinds of homomorphic encryption: Partial and store them. Section 2 in the paper explains the
Fully homomorphic schemes. related work. Section 3 is based on the research
● Partial homomorphic scheme allow some methodology followed by Section 4 where we tell
only some computations to be performed about the proposed model. Lastly, Section 5 is
on ciphertext like addition, multiplication. about the results and future scope.
● In Fully homomorphic scheme most of the
operations can be carried out on the
ciphertext and was developed by Craig
2. RELATED WORK
Gentry.
Homomorphic encryption is a type of encryption
Jain, R., Madan, S., Garg, B., Kapila, Y., &
methodology which allows operations to be made
Gupta, A [1] analyzes Paillier algorithm and is
in encrypted text, thereby generating an encrypted
used to perform addition of encrypted numbers
result that, once decrypted, matches the result
and subsequently perform decryption.
operations performed in the plain text.
Homomorphic encryption is used in many
Tebaa, M., & Hajji, S. E. (2014) [2] studies the
modern communication architectures. Other
function of various Homomorphic Encryption
secure homomorphic systems, such as the secure
algorithms such as Goldwasser-Micali, Paillier,
voting system and secure information retrieval
El Gamal, RSA, Gentry and Boneh-Goh-Nissim.
schemes. It also helps secure distributed
The evaluation is based on four individualities;
computing. Paillier cryptosystem is one such
type of homomorphic encryption, keys used,
probabilistic asymmetrical algorithm for public
security used for and data privacy.
key cryptography. Paillier algorithm is an
additive homomorphic encryption, where we can Zhao, F., Li, C., & Liu, C. F. (2014, February) [3]
compute the encryption of n1+n2 given only the talks about the security issues as well as privacy
public key and the encryption of n1 and n2. As concerns related to cloud computing like the role
discussed earlier that in this paper we will of various cloud computing models, data
implement e-voting or electronic voting using the integrity, availability and confidentiality in
paillier cryptosystem. Electronic voting gives us fulfilling these concerns. It studies a fully
the advantage to cast votes from any remote place homomorphic encryption algorithm proposed by
through the use of internet. Users can open the Craig Gentry and verifies the additive and
voting portal and cast their votes sitting in the multiplicative properties of fully homomorphic
comfort of their homes and offices. encryption technique.
Requirements for e-voting system: Van Dijk et. al., (2010, May) [4] talks about fully
1. Authentication: Just the authorized users homomorphic encryption algorithm which makes
should be able to vote. use of simple integer arithmetic. The principal
2. Unique vote: All the votes should be open problem is to increase the encryption
unique and should not be casted more than efficiency.
once.
Dhote, C. A. (2016) [5] focuses on data storage in
3. Result Accuracy: The system should be encrypted format stored in plaintext using fully
able to generate the results with great homomorphic encryption algorithm. The aim of
accuracy.

HTTPS://[Link]/LINK/[Link] ELSEVIER-SSRN (ISSN: 1556-5068) 97

Electronic copy available at: [Link]


Proceedings of 4th International Conference on Computer and Management ICCM (2018)

this paper is to execute actions on data which is cryptographic tools used in designing e-voting
encrypted without decrypting it. system.

Cheon, J. H., Jeong, J., Lee, J., & Lee, K. (2017,


April) [6] provided a predictive analysis on
encrypted medical data by using homomorphic
encryption. They approximated their data using 3. RESEARCH METHODOLOGY
Non-Adjacent form encoding and minimax
approximation (earlier done through Tayor Paillier is a type of encryption based on a key
Approximation) which removed the input barrier. pair. In paillier’s encryption every user gets a
public and a separate private key, while the
Dyer, J., Dyer, M., & Xu, J. (2017) [7] messages are encrypted with their public key at
proposes a useful homomorphic encryption the same time their private key is used to decrypt
over the integers. arXiv preprint the same message.
arXiv:1702.07588 analyzed four
homomorphic encryption algorithms on the The Paillier encryption system has the built-in
basis of computational speed, security and feature that a given plaintext can contain many
demonstrates the practicality of each algorithm different encrypted texts, making it much more
robust "ready to use" for a variety of
Damgård, I., & Jurik, M. (2001, February) [8] cryptographic attacks. This is due to the addition
A generality, a simplification and some of additional randomness in the encryption, which
functions of paillier's probabilistic public-key can be eliminated with the key due to the
technique. Springer, Berlin, Heidelberg quadratic residency property. In the
suggested a generalization of paillier’s implementation, a similar effect can be achieved
probabilistic key technique in International in RSA through the padding or other schemes, but
Workshop on Public Key Cryptography (pp. in Pallier there is the version "textbook" in the
119-136). whose expansion factor has been base of the algorithm, which is not the case of the
reduced resulting in reduced block size and at "book version" of text "RSA.
the same time holding homomorphic property.
It also shows how these can be allowed to 𝐸𝑛𝑐(𝑚) = 𝑔𝑚𝑟𝑛 𝑚𝑜𝑑 𝑛2
efficient e-voting systems.
To add two encrypted numbers, following
Pan, M., Sun, J. and Fang, Y. (2011) [9] formula can be used:-
Backstage Cleansing: Secure Spectrum
Auction with a Paillier Cryptosystem. The 𝐷 (𝐸(𝑚1, 𝑟1). 𝐸(𝑚2, 𝑟2)𝑚𝑜𝑑𝑛2) = 𝑚1 +
IEEE Journal on Certain Areas of 𝑚2𝑚𝑜𝑑𝑛
Communications, 29 (4), 866-876, proposes a
THEMIS system, a secure frequency auction 𝐷(𝐸(𝑚1 , 𝑟1 ). 𝑔𝑚2 𝑚𝑜𝑑𝑛2 ) = 𝑚1 +
using the Paillier cryptosystem, a fraudulent 𝑚2 𝑚𝑜𝑑𝑛
auctioneer fraud and offers between bidders
and to prevent the bid-rigging between the
auctioneer and bidder.
4. PROPOSED MODEL
Wang, K. H., Mondal, S. K., Chan, K., & Xie,
X. (2017) [10] reviews all the challenges that We have researched about various cryptosystems
are faced by modern e-voting system. They and chose paillier as a base to implement the e-
provide a vast amount of security requirements voting system. Our basic aim is to compare
that allows researcher to design an e-voting various cryptosystems and their behaviors while
system. They also summarize different encrypting and decrypting data given a public and

HTTPS://[Link]/LINK/[Link] ELSEVIER-SSRN (ISSN: 1556-5068) 98

Electronic copy available at: [Link]


[Type here]

private key. On top of that we have built a


website which provides us a UI to create
elections, specify candidates and able to store the
votes in encrypted format, without anyone
knowing, which voter voted to whom.

Figure 3. Code for encryption, decryption and key generation

3) It was required that each candidate was


given a unique number which would be added
at the end.
The following formula can be used to generate
unique numbers: -
10^(log10(number of voters)*i) where i is the
candidate number (serial wise).
Figure 1. Flowchart depicting user authentication

1) The vote table stores the votes for each


candidate in encrypted format in attribute
‘encrypted vote’.

Figure 2. Example of an encrypted vote

2) The framework we used to make the


website is rails which uses ruby language. We
implemented paillier cryptosystem in ruby
language to encrypt, decrypt and add (already
encrypted) data.

Figure 4. Flow chart depicting voting mechanism

5. RESULTS

HTTPS://[Link]/LINK/[Link] ELSEVIER-SSRN (ISSN: 1556-5068) 99

Electronic copy available at: [Link]


Proceedings of 4th International Conference on Computer and Management ICCM (2018)

In the proposed e-voting system we have constantly. We studied about the various
implemented and created a Ruby library for homomorphic encryption cryptosystems and
Paillier algorithm which has different proposed a secure e-voting system using
functionalities like encryption of votes, homomorphic encryption (paillier
decryption of total votes casted and addition of cryptosystem) which is available is over the
different encrypted votes to the same cloud. In future a more robust and efficient
candidate. This e-voting system is remotely key generating mechanism could be adopted
available to the users and follow the security such as by studying octonions algebra as the
standards enforced by homomorphic current key generation function generates a
encryption where votes after casting are very long key because of which it takes more
encrypted and stored using paillier’s time to evaluate the final tally. It would
algorithm. We have created a GUI for the e- generate a smaller key and thus would reduce
voting platform where users can interact with the total time while tallying, if octonions
the system easily and securely. The GUI algebra is implemented.
provides an option for admin dashboard where
the admin can hold elections and generate a
tally of results after the elections are over.
7. REFERENCES

[1] Jain, R., Madan, S., Garg, B., Kapila, Y.,


& Gupta, A. (2017). E-Voting System
using Homomorphic Encryption in a
Cloud Based Environment. International
Journal of Security and Its Applications,
11(5), 59-68.
[2] Tebaa, M., & Hajji, S. E. (2014). Secure
cloud computing through homomorphic
encryption. arXiv preprint
Figure 5. User Dashboard
arXiv:1409.0829.
[3] Zhao, F., Li, C., & Liu, C. F. (2014,
February). A cloud computing security
solution based on fully homomorphic
encryption. In Advanced Communication
Technology (ICACT), 2014 16th
International Conference on (pp. 485-
488). IEEE.
[4] Van Dijk, M., Gentry, C., Halevi, S., &
Vaikuntanathan, V. (2010, May). Fully
homomorphic encryption over the
integers. In Annual International
Conference on the Theory and
Figure 6. Admin Dashboard depicting results of an election Applications of Cryptographic
Techniques (pp. 24-43). Springer, Berlin,
Heidelberg.
[5] Dhote, C. A. (2016). Homomorphic
6. CONCLUSION AND FUTURE SCOPE
encryption for security of cloud data.
We have seen why security and privacy play Procedia Computer Science, 79, 175-181.
such an important role in cloud computing and [6] Coron, J. S., Mandal, A., Naccache, D., &
that cloud computing market is booming Tibouchi, M. (2011, August). Fully

HTTPS://[Link]/LINK/[Link] ELSEVIER-SSRN (ISSN: 1556-5068) 100

Electronic copy available at: [Link]


[Type here]

homomorphic encryption over the Public Key Cryptography (pp. 420-443).


integers with shorter public keys. In Springer, Berlin, Heidelberg.
Annual Cryptology Conference (pp. 487- [16] Pulier, E., Martinez, F., & Hill, D. C.
504). Springer, Berlin, Heidelberg. (2015). U.S. Patent No. 8,931,038.
[7] Tebaa, M., & Hajji, S. E. (2014). Secure Washington, DC: U.S. Patent and
cloud computing through homomorphic Trademark Office.
encryption. arXiv preprint [17] Hirt, M., & Sako, K. (2000, May).
arXiv:1409.0829. Efficient receipt-free voting based on
[8] Mohanta, B. K., & Gountia, D. (2013). homomorphic encryption. In
Fully homomorphic encryption equating International Conference on the Theory
to cloud security: an approach. IOSR and Applications of Cryptographic
Journal of Computer Engineering (IOSR- Techniques (pp. 539-556). Springer,
JCE) Volume, 9. Berlin, Heidelberg.
[9] Negi, A., & Goyal, A. (2018). Optimizing [18] Madise, Ü., & Martens, T. (2006). E-
Fully Homomorphic Encryption voting in Estonia 2005. The first practice
Algorithm using RSA and Diffie-Hellman of country-wide binding Internet voting in
Approach in Cloud Computing. the world. Electronic voting, 86(2006).
[10] Gentry, C., Sahai, A., & Waters, B. [19] Paillier, P. (1999, May). Public-key
(2013). Homomorphic encryption from cryptosystems based on composite degree
learning with errors: Conceptually- residuosity classes. In International
simpler, asymptotically-faster, attribute- Conference on the Theory and
based. In Advances in Cryptology– Applications of Cryptographic
CRYPTO 2013 (pp. 75-92). Springer, Techniques (pp. 223-238). Springer,
Berlin, Heidelberg. Berlin, Heidelberg.
[11] Neff, C. A. (2001, November). A [20] Pan, M., Sun, J., & Fang, Y. (2011).
verifiable secret shuffle and its Purging the back-room dealing: Secure
application to e-voting. In Proceedings of spectrum auction leveraging paillier
the 8th ACM conference on Computer cryptosystem. IEEE Journal on Selected
and Communications Security (pp. 116- Areas in Communications, 29(4), 866-
125). ACM. [Link], J. H., Jeong, J., Lee, J., &
[12] Maaten, E. (2004). Towards remote e- Lee, K. (2017, April). Privacy-preserving
voting: Estonian case. Electronic Voting computations of predictive medical
in Europe-Technology, Law, Politics and models with minimax approximation and
Society, 47, 83-100. non-adjacent form. In International
[13] Gritzalis, D. A. (2002). Principles and Conference on Financial Cryptography
requirements for a secure e-voting and Data Security (pp. 53-74). Springer,
system. Computers & Security, 21(6), Cham.
539-556. [21] Jiang, L., Xu, C., Wang, X., & Lin, C.
[14] Horváth, M. (2015, January). Attribute- (2017). Statistical learning based fully
based encryption optimized for cloud homomorphic encryption on encrypted
computing. In International Conference data. Soft Computing, 21(24), 7473-7483.
on Current Trends in Theory and Practice [22] Chervyakov, N., Babenko, M.,
of Informatics (pp. 566-577). Springer, Tchernykh, A., Kucherov, N., Miranda-
Berlin, Heidelberg. López, V., & Cortés-Mendoza, J. M.
[15] Smart, N. P., & Vercauteren, F. (2010, (2017). AR-RRNS: Configurable reliable
May). Fully homomorphic encryption distributed data storage systems for
with relatively small key and ciphertext Internet of Things to ensure
sizes. In International Workshop on security. Future Generation Computer
Systems.

HTTPS://[Link]/LINK/[Link] ELSEVIER-SSRN (ISSN: 1556-5068) 101

Electronic copy available at: [Link]


Proceedings of 4th International Conference on Computer and Management ICCM (2018)

[23] Dyer, J., Dyer, M., & Xu, J. (2017).


Practical homomorphic encryption over
the integers. arXiv preprint
arXiv:1702.07588.
[24] Damgård, I., & Jurik, M. (2001,
February). A generalisation, a simpli.
cation and some applications of paillier's
probabilistic public-key system.
In International Workshop on Public Key
Cryptography (pp. 119-136). Springer,
Berlin, Heidelberg.
[25] Wang, K. H., Mondal, S. K., Chan, K., &
Xie, X. (2017). A review of contemporary
e-voting: Requirements, technology,
systems and usability. Data Science and
Pattern Recognition, 1(1), 31-47.

HTTPS://[Link]/LINK/[Link] ELSEVIER-SSRN (ISSN: 1556-5068) 102

Electronic copy available at: [Link]

You might also like