INDIAN INSTITUTE OF INFORMATION TECHNOLOGY

POM-RESEARCH PAPER

DIGITAL VOTING USING BLOCKCHAIN

GROUP MEMBERS:

S.no Names Roll Number

1. Anuj Mittal IIT2018063

2. Aakash Mehta IIT2018064

3. Agamjot Singh IIT2018060

4. Aditya Kaithwas IIT2018061

5. Suneet Makkar IIT2018090

6. Kisalaya Kishore IIT2018079

1
 CONTENTS

1. INTRODUCTION TO BLOCKCHAIN
1.1. ORIGIN
1.2. EMPLOYMENT OF BLOCKCHAIN IN VARIOUS SECTORS

2. NEED OF BLOCKCHAIN TECHNOLOGY

2.1. PECULIARITIES IN CURRENT VOTING SYSTEM
2.2. HOW BLOCKCHAIN CAN CHANGE HOW WE VOTE

3. ARCHITECTURE AND DESIGN OF PROPOSED E-VOTING SYSTEM

3.1. REGISTRATION OF ELIGIBLE VOTERS
3.2. PRE-VOTING PHASE
3.3. VOTING PHASE
3.4. ANALYSIS OF THE DESIGN

4. IMPROVEMENT AND EXTENSIONS

5. CONCLUSION

6. REFERENCE

2
 1. INTRODUCTION TO BLOCKCHAIN

1.1 ORIGIN

First ever work on cryptographically secured chain of blocks was described by Stuart Haber
and W.Scott Stornetta in 1991[1].
Blockchain was first conceptualized by Satoshi Nakamoto in 2008[2].

WHAT IS BLOCKCHAIN?

Blockchain is a distributed database of records or public ledger of all the digital events or
transactions that have been shared among concerned parties. Each transaction in the public
ledger is verified by consensus of a majority of the participants in the system. And once entered
information can never be erased.

The main hypothesis is that the blockchain establishes a system of creating a distributed
consensus in the digital online world. This allows to differentiate between public ledger record
and irrefutable record. It opens the door to a democratic open and centralized system. There are
tremendous opportunities in this technology and revolution in this space has just begun.

Each block contains a cryptographic hash of the previous block, a timestamp, and transaction
data[3].

3
1.2 EMPLOYMENT OF BLOCKCHAIN IN VARIOUS SECTORS

BANKING
Banks serve as the critical storehouses and transfer hubs of value. As a digitized, secure, and
tamper-proof ledger, blockchain could serve the same function, injecting enhanced accuracy and
information-sharing into the financial services ecosystem.

CLOUD STORAGE
Enterprises that offer cloud storage often secure customers’ data in a centralized server, which
can mean increased network vulnerability from attacks by hackers. Blockchain cloud storage
solutions allow storage to be decentralized — and therefore less prone to attacks that can cause
systemic damage and widespread data loss.

STOCK TRADING
For years companies have worked to ease the process of buying, selling, and trading stocks, and
now new blockchain-focused startups are looking to automate and secure the process more
efficiently than any past solution.

GOVERNMENT AND PUBLIC RECORDS

The management of public services is yet another area where blockchain can help lessen paper-
based processes, minimize fraud, and increase accountability between authorities and those they
serve.

4
 2. NEED OF BLOCKCHAIN TECHNOLOGY

2.1 PECULIARITIES OF CURRENT VOTING SYSTEM

Tampering with the voting machines:

There are many frauds which threatens every voting systems like tampering with the
software of the electronic machine altering vote totals and being biased towards a
particular candidate[4].
Destruction of ballots:
Electoral fraud is also conducted by destroying ballots of an expected majority area of
opposing candidate or party thereby affecting the whole result.
Criminalism during voting:
Polling places in an area known to support a particular party or candidates may be
targeted for vandalism, threats or destruction. Thugs hired by candidates try to
manipulate the minds of common people by adopting unhealthy tactics.
Time consuming:
The current voting system is very time consuming as people have to go to polling places to
vote for parties and people have to wait for their turns to vote.
Transportation:
In conventional method of voting, voting machines are transported to a particular place
where the counting of votes is held. Transportation costs a lot of money, vulnerability to attacks
or accidents plus it is time consuming.
What will it take to make the largest democracy of the world also the greatest? What matters
most is that voting systems ensure the will and betterment of the voters and the legitimacy of the
leaders.With the peculiarities in the current voting systems increasingly becoming exposed, the
time to look at alternative models has come. Now that the parliamentary committee has set in
motion this serious debate, one hopes that the voting system itself would be taken up as a key
reform. [5] So here comes a new blockchain technology to change the systems of voting so that
the country has right people forming the government.

5
2.2 HOW BLOCKCHAIN TECHNOLOGY CAN CHANGE THE WAY WE
VOTE ?

In a time of technical evolution, there is not a better method than blockchain technology. This
method provided by blockchain technology is more secure, easier and allows more people to
perform their fundamental civic duty [6].It could result in the establishment of a sufficiently
secure system that could itself prepare accounts for all voters and give them an option to vote
through a platform of their choice. If they still want to go into a polling station, they could do so.
But they could also use a government-approved website or smartphone that would access and
record data through the blockchain.

One of the biggest revolutions that would be introduced by blockchain is its decentralized
system. That's good news when it comes to democracy because the more transparency there is,
the better it is. It's also much more safe and solid than many other alternatives such as paper-
based systems or other platforms which are digital in nature. In fact, it is secure by design. This
combination of security, transparency and decentralization is exactly why blockchain technology
is such an innovative idea. For example, it could be used in the real estate industry to create
digital records for every property to track ownership, maintenance records and other information
about every house in the country.

A 2018 election in Sierra Leone banked heavily on this blockchain technology with
approximately 70% of votes being stored and verified with the blockchain. It was likely the first
of its kind, especially at such a colossal scale. What’s interesting about this is that blockchain
could actually minimize the amount of corruption within the system whilst simultaneously
enabling the results to be evaluated in a reduced time[7]. There would be no need for prolonged
counting and recounting processes. The data would be accessible by everyone. Pretty much any
relevant party could "review, count and validate" the data, ultimately allowing citizens to hold
their governments to more accountability than ever before.

The most influential quote of all the variety of quotes is from capitalist who said, "The
blockchain does one thing: It replaces third-party trust with mathematical proof that something
happened. And when it comes to counting and processing votes, I'm pretty sure we'd all prefer
mathematical proof than to take something on trust” [8] .By now, you should have acquired a good
idea of why blockchain technology has the potential to revolutionize the procedure of our
voting . The only real question is when blockchain-based voting will become the new norm
instead of just an exception. It can only be a matter of time.

6
 3. ARCHITECTURE AND DESIGN OF PROPOSED E-VOTING

SYSTEM

3.1 REGISTRATION(Through Aadhar card/Biometrics)

First a voter is verified for establishing security in the system. Making sure that someone’s
identity isn’t being misused for fraudulent purposes is important, especially when voting is
considered, where every vote matter. To allow users to register to vote our proposed service
utilizes both postal based forms as well as web forms requiring the same information to ensure
we cater for those without a direct internet connection. This information includes their national
identity number (Aadhar), postal address, Biometric, optional email address and a password[9].
All of this information then form a transaction for the user agreeing with the government that
they are asking to vote.
Once someone has registered an automated government miner analyses the transaction and if
they haven’t been awarded or denied a vote the miner will make the decision as to whether to
verify the user or not. If the user is verified, they will be sent a ballot card with their information
on it to both their home address and email address if provided. They will also be sent a randomly
generated password to use on the polling stations. Once this correspondence has been sent, the
miner will create a transaction giving the user a vote from an infinite government pool of votes
on the voter blockchain[10].
During this process, a voter blockchain is used to keep a record of both transactions taking place
at each stage of this process for each voter:
1. Firstly, a transaction is created when a user ‘registers’.
2. The next transaction is created when a government miner authorizes that user’s right to vote.
After the correspondence is received by the user they can then await voting to open to use their
credentials to vote. It is important to note that this voter blockchain will never contain details of
the vote cast by the user.

7
3.2 PRE-VOTING PHASE:

When deciding on the architecture we took strong inspiration from both the distribution and
availability of the Bitcoin network and the aggregation process of traditional voting. This
network is a multi-tiered, decentralised infrastructure in which houses are the two distinct
blockchains, the network[11] is divided into three abstract tiers, National, Constituency and Local
blockchains.
The local tier contains all the digital polling stations across the country, each of which is
associated to a constituency node. A local node is setup to only communicate with the other local
nodes under the associated constituency node and the constituency node itself.

The constituency tier consists of all the nodes that are deemed to be at a constituency level.
These nodes would be directly connected to each other and to a subset of polling stations
depending on their location. The national tier is a collection of nodes that are not tied to location,
their pure purpose is to mine transactions and add blocks to the vote blockchain, all constituency
nodes communicate to a national node and national nodes can communicate with each other[12
Independent bodies will monitor and audit the voting process. These bodies will host or have
access to a national node and will be able to verify that the unencrypted data matching the
encrypted votes. Individuals and organisations can become a national node. These applications
are processed by the government to ensure that they meet the minimum requirements set by the
governing body. These individuals will also act as miners during the counting process.

8
3.3 THE VOTING PHASE:

When it is time to vote, authentication of a user requires three distinct pieces of evidence; their
identification number, the password supplied on registration, their ballot card which contains a
QR code. As these are the two methods of voting the way the user will input the authentication
details shall be different; however, to vote they are required to provide all three pieces of
information. It is also important to note that each user has been registered at a certain
constituency so that they will only be able to vote at a local polling station within that
constituency or via the internet at the URL provided on the ballot card. Voting station will
conduct the survey accordingly and carry on.
Voting station will also consult the voter blockchain to ensure that voter has not already used up
their vote. If the user does have a vote, then the station will then allow the user to continue to the
voting screen. If not, then system will respond to the user appropriately.

After selecting their vote and then confirming the submission, the vote will become a transaction,
it will be encrypted with the relevant constituencies public key. This transaction is then passed to
the constituency node where it is added to a block and the update is then pushed to all other
nodes connected to that particular constituency node. The connected nodes then pass the data on
to their peers until the whole network is updated[13].

9
3.4 ANALYSIS OF THE DESIGN:

Within our proposal we have tried to design a service and system that minimises the size of
attack vectors to prevent potential malicious attacks. We have tried to evaluate and analyse our
design from various perspectives to make sure we have thought about each step of the voting
process. This section of the report discusses the potential risks associated with our proposal and
suggests actions that can be taken to help mitigate them[14].

One risk is if a voter were to forget their ID, password or polling card on the day of voting. In
this case the voter will be unable to cast their vote as they cannot enter the system. Possible risk
mitigations include the voter returning later that day with the correct information or the
implementation of a backup authentication service such as by phone. Alternatively, a forgotten
password system could be added to the voter registration website[15] this could work in much the
same way as recovering a password works on other websites. However, this increases the risk of
a hacker attempting to change a voter’s password without their knowing.

A 51% attack is a potential threat to our proposed design. The basis of the attack being that
someone could theoretically control a majority of the digital voting mining hash-rate, leading to
them being able to manipulate the public ledger. The chances of this type of attack occurring are
slim due to the immense cost needed to purchase hardware capable of this scale of processing.
We also have the added security of an auditor who checks and keeps track of people connecting
to the network and the locations of each node[16]. This is a feature that current systems such as
bitcoin lack.

The online aspect of the voting within our system is the largest attack vector for hackers as they
could potentially exploit voters through their own devices in a host of ways. To combat this
software could be developed that could be downloaded onto the clients device to establish a
secure connection to the polling station.

10
4. IMPROVEMENTS AND EXTENSIONS:

In this section, we discuss some possible further improvements and extensions when applying
the e-voting protocol in special elections and scenarios.

Privacy of Data Transmission In our protocol, the communication through the blockchain
network may divulge voters’ IP addresses, which may lead to the exposure of connections
between voters and ballots via network analysis. To enhance voters’ privacy, we recommend
voters to use anonymity services like proxies with which voters can hide their IP addresses[17].

Data Confidentiality and Neutrality According to our protocol, because of the transparency
property from blockchain, ballots are visible when they are cast to the blockchain network. This
exposes the progress of the election during the voting phase, and may greatly influence the
outcome of the election. Here, we provide two possible solutions for this problem. A direct
solution is to control the access of blockchain, by simply employing a permissioned blockchain
for the election. The permissioned blockchain is more flexible and there exist several promising
solutions of access control. However, providing certain extent of data confidentiality,
transparency is somehow lost. To keep transparency, we may also introduce a ballot encryption
mechanism here. The basic idea is: voters encrypt the message Ballot using one new public 10 Yi
Liu and Qi Wang key provided by organizer, and the organizer open the corresponding private
key for ballot decryption before the post-voting phase. The election then becomes confidential,
and all ballots are enclosed until the end of the voting phase.

Dishonest behaviours from the Organizer and Inspectors Corruption may happen if the organizer
and inspector conspire together, since both of their signatures are components of a valid ballot.
To avoid this kind of dishonest behaviours, we can introduce more inspectors such that the
corruption cost is greatly increased.

11
5. CONCLUSION:
To close, our service proposal comprises of a geographically distributed network comprising of
machines from both government and public infrastructure; this infrastructure houses two
distinctly separate blockchains, one for voter information such as who has voted and the other for
vote information such as what has been voted. These blockchains are held completely separately
to remove any threat to link votes for certain parties back to individual voters while maintaining
the ability to track who has voted and how many votes are actually present.

The blockchain containing information of who has registered to vote also allows our service to
ensure each voter is unique. Once registered you are then allocated a vote after verification of
your details has been completed. To ensure these registered voters are who they say they are
when voting begins there is a 3 factor authentication method. Further to this we also need to
ensure they are not forced to vote in a particular way so we have incorporated a double-check
service where by users shall be prompted a second time to confirm their submission before the
vote is sent; this also then allows us to almost eradicate accidental votes[18].

Also, due to the encryption mechanism we are using it would be close to impossible for any
person(s) to gain access to all the votes without first taking control of the entire service network.
Moving on from this the publication method of the private keys allows anyone to read the
blockchain of votes and decrypt them with the newly available constituency private keys to
verify the result of the election.

12
 REFERENCES

[1] “How to time-stamp a digital document (January 1991)”.

[2] The Economist (31 October,2015).

[3] Narayanan Arvind: Goldfeder, Steven (2016)].

[4] Mark Williard (2014).

[5] http://www.insightsonindia.com/2008/07/25/electoral-system-in-india-flaws-and-reforms/.

[6] https://www.thebalance.com.

[7] https://www.inc.com/james-paine/how-blockchain-could-change-how-we-vote.html.

[8] American venture capitalist, Adam Draper.

[9] https://e-estonia.com/component/electronic-id-card/.

[10]http://mfa.ee/sites/default/files/contenteditors/2015%20Parliamentary%20elections%20Inter
net%20voting%20system.pdf.

[12] https://www.economist.com/sites/default/files/plymouth.pdf.

[13] https://www.quora.com/Can-blockchain-technology-be-used-in-securing-digital-voting.

[14] https://techcrunch.com/2018/03/14/sierra-leone-just-ran-the-first-blockchain-based-election/

[15] Fouard, L., Duclos, M., Lafourcade, P.: Survey on electronic voting schemes. supported by
the ANR project AVOTE (2007)

[16] Gritzalis, D.A.: Secure electronic voting. Volume 7. Springer Science & Business Media
(2012)

[17] Carter, L., B´elanger, F.: Internet voting and political participation: an empirical comparison
of technological and political factors. DATA BASE 43(3) (2012) 26–46.

[18] Madise, U., Martens, T.: E-voting in estonia 2005. the first practice of country- ¨ wide
binding internet voting in the world. In Krimmer, R., ed.: Electronic Voting 2006.

13