AUTHENTICATION FOR REMOTE VOTING USING VISUAL CRYPTOGRAPHY full report
Active In SP
Joined: Apr 2010
06-06-2010, 01:36 PM
AUTHENTICATION FOR REMOTE VOTING USING VISUAL CRYPTOGRAPHY.docx (Size: 125.34 KB / Downloads: 176)
AUTHENTICATION FOR REMOTE VOTING USING VISUAL CRYPTOGRAPHY
Ms. Manisha H. Shasri
Ms. Pallavi V. Chavan
Mr. R. S. Mangrulkar
Trustworthy elections are essential to democracy. Elections are complex and involve many components including voter registration, ballot preparation and distribution, voter authentication, vote casting, tabulation, result reporting, auditing, and validation. In this paper, we only considered voter authentication using visual cryptography. The system proposed here uses visual cryptography to provide mutual authentication for voters & election servers. Internet based voting offers many benefits including low cost & increased voter participation. Remote Voting system considers security & human factors carefully and in particular make sure that they provide voters with reliable and intuitive indications of the validity of the voting process. We realize there are many other important security issues to address before Internet voting could be adopted in governmental elections such as database security and denial-of-service attacks on the Internet. Using remote voting the voter cannot only vote from his poll site but also from any remote place. The authentication provided by this scheme is very secure as compare to other schemes. emote voting through Internet is not currently used in India. To simplify job of election officers and to provide fast election process, the remote voting systems are evolved. In previous poll site voting system there was no security for voter authentication. With traditional poll site voting, voters authenticate themselves by providing identification or an affirmation to a trusted poll worker; a poll site authenticates itself to a voter by being at a well-publicized physical location and having officials representing several different organizations present Due to this voting selling was easy also the voting process was very complex. Remote voting not only provides fast voting process but also it is suitable for the people of India to vote from other countries Main aim of remote voting system is to provide complete privacy to the voter and the best integration of the voting system. The core concentration of this paper is voter authentication. The voter need not understand how visual cryptography works but being directly involved in visual cryptography. As visual cryptography does not need any computations for decoding purpose, the task for the voter is easy. visual cryptography encrypts the information in such a way that decryption can be done without using any mathematical computations. Still some of the threats are possible to this approach, which are as, Vote Coercion, Vote Selling, Vote Solicitation, Online Registration ,Voter Privacy, Could have a scrawny teenage script kiddy but now a foreign government.
Voter Authenticity: Ensure that the voter must identify himself (with respect to the registration database) to be entitled to vote.
Registration: The voter registration shall be done in person only. However, the computerized registration database shall be made available to polling booths all around the nation.
Voter Anonymity: Ensure that votes must not be associated with voter identity.
System Integrity: Ensure that the system cannot be reÃ‚Â¬configured during operation.
Data Integrity: Ensure that each vote is recorded as intended and cannot be tampered with in any manner, once recorded (i.e., votes should not be modified, forged or deleted without detection).
Secrecy / Privacy: No one should be able to determine how any individual voted.
Reliability: Election systems should work robustly, without loss of any votes, even in the face of numerous failures, including failures of voting machines and total loss of network communication. The system shall be developed in a manner that ensures there are no malicious codes or bugs.
Availability: Ensure that system is protected against accidental and malicious denial of service attacks. Also, setup redundant communication paths so that availability is ensured.
Simplicity: The system shall be designed to be extremely simple, as complexity is the enemy of security. System Accountability: Ensure that system operations are logged and audited.
Distribution of Authority: The administrative authority shall not rest with a single entity. The authority shall be distributed among multiple administrators, who are known not to collude among themselves (e.g., different political parties).
Many approaches are there to authenticate to the voting system as authentication by password, authentication using mechanical key or using some mechanical factors. The approach used here is based on visual cryptography. A remote authentication is provided for both voters & voting system using visual cryptography. Using Visual Cryptography it is possible to establish authentication that is resistant to large-scale frauds.
ASSUMPTION: Election office mails each voter a voting packet including a printed transparency. The voter will use this transparency to authenticate himself to voting system. Establishing trust is one of the most important human and human-computer interactions. Internet-based voting offers great convenience, but does not offer such obvious authentication methods. Remote voting in governmental elections is done through absentee ballots that offer little security, and are slow and expensive to tabulate. Voters vote with their feet (and votes), of course, and remote voting is becoming increasingly accepted and popular The amount of trust needed before a voter is willing to cast a vote depends on the voter and the election. In typical United States elections, most voters are willing to place a high amount of confidence in the integrity of the process without needing definite mechanisms to guarantee their vote and identity cannot be linked. In other countries voters may not have such confidence. Iraq has declared Saddam Hussein the winner with 100 percent of the votes in a referendum granting him another seven-year term, bringing bursts of celebratory gunfire in Baghdad's streets. This is probably one of the reasons why every one of the 11,445,638 eligible voters voted "Yes" for Saddam Hussein in Iraq's 2002 presidential referendum (the vote recording process is also suspect). All 11,445,638 Iraqis eligible to vote had done so and every single one of them answered "Yes" to another seven-year term for Saddam, 65, who was appointed president in 1979. Iraqi officials said popular outrage at American threats to Saddam's regime made the turnout and percentage even higher than the last vote, in 1995, when Saddam received a 99.96 "yes" vote. This result is real. With traditional voting, most voters are willing to accept this because they believe the poll workers are trustworthy. With remote voting, stronger measures are required, and it is important that the existence of those measures is clearly conveyed to voters in a way that establishes an appropriate level of trust.
We propose to provide remote authentication for both voters and voting systems using visual cryptography. David Chaum first proposed applying visual cryptography to elections to allow voters to verify their votes are included correctly in the final tabulation. Using Chaum's secret-ballot receipts, a machine prints a receipt showing the results of a cast vote. The voter chooses to keep the top or bottom layer, each being unreadable without the other layer. Upon leaving the polling place, each voter can check to make sure the layer is correct and the vote was counted by visiting an official website that has a listing of all voter receipts used for tabulation. By itself, the voter's receipt does not reveal any information about the actual vote. We do not consider ways for voters to verify their vote was recorded correctly in the final tabulation, but rather focus on the authentication process. By using visual cryptography, we believe it will be possible to establish authentication in a way that is satisfying to voters and resistant to large-scale fraud. Our authentication scheme assumes the elections office mails each voter a voting packet including a printed transparency. This would not dramatically increase the cost of conducting an election, since many jurisdictions (including the state of Washington) already mail any voter requesting a packet before each election. The voter will use this transparency to authenticate herself to the voting system and to verify that the election server is legitimate.
The simplest form of visual cryptography separates an image into two layers so that either layer by itself conveys no information, but when the layers are combined the image is revealed. Figure 4.1 illustrates how an image is divided into layers. One layer can be printed on a transparency, and the other layer displayed on a monitor. When the transparency is placed on top of the monitor and aligned correctly, the image is revealed. For each image pixel, one of the two encoding options is randomly selected with equal probability. Then, the appropriate colorings of the transparency and screen squares are determined based on the color of the pixel in the image.
This scheme provides theoretically perfect secrecy. An attacker who obtains either the transparency image or the screen image obtains no information at all about the encoded image since a black-white square on either image is equally likely to encode a clear or dark square in the original image. Another valuable property of visual cryptography is that we can create the second layer after distributing the first layer to produce any image we want. Given a known transparency image, we can select a screen image by choosing the appropriate squares to produce the desired image. One of the most obvious limitation of using visual cryptography in the past was the problem of the decoded image containing an overall gray effect due to the leftover black sub pixel from encoding. This occurred because the decoded image is not an exact preproduction, but an expansion of the original, with extra black pixel. Black pixel in the original document remain black pixel in the decoded version, but White pixel becomes gray. This resulted in a lot of contrast to the entire image. The extra black sub pixel in the image cause the image to become distorted.
D - Secret information. K - number of shares generated from D. share - piece of information.
Divide data D into n pieces in such a way that D is easily reconstructable from any k pieces, but even complete knowledge of any k-1 pieces reveals no information about D. Stacking two pixels (each consists of four sub-pixels) can occur for example the following two cases:
Secret sharing scheme is a method of sharing secret information among a group of participants. In a secret sharing scheme, each participant gets a piece of secret information, called a share. When the allowed coalitions of the participants pool their shares, they can recover the
shared secret; on the other hand, any other subsets, namely non-allowed coalitions, can not recover the secret image by pooling their shares. In the last decade, various secret sharing schemes were proposed, but most of them need a lot of computations to decode the shared secret information.
The basic 2 out of 2 visual cryptography model consist of secret message encoded into two transparencies , one transparency representing the cipher text and the other acting as a secret key. Both transparencies appear to be random dot when inspected individually and provide no information about the original clear text. However, by carefully aligning the transparencies , the original secret message is reproduced. The actual decoding is accomplished by the human visual system. The original is encrypted into 2 transparencies You need both transparencies to decode the message. An example is given below.
A voter visits the election web site and enters the typeable version of the key Ki found on the transparency. We can encode a 64-bit key in 12 characters selected from lowercase letters and numbers. Many software packages require much longer input strings for their installation, so voters should not mind typing 12 characters. The election web site maintains a list of the Ki values used to generate the transparencies and checks that the entered key is on the list and has not been used already (extensions that would allow a voter to change a previously cast vote are possible but not considered here). If the entered Ki is valid, the election server (which has access to Kg) can calculate the corresponding transparency image. The election server then generates a random string to use as a password, and generates an image containing that string rendered as a bitmap image. The complementary image to the password image for the voter's transparency is generated and displayed on a web page returned to the voter. After the web server displays the corresponding image generated from Kg, the voter holds the transparency up to the screen to reveal the password (see Figures ). To continue the voting process, the voter enters the revealed password. This protocol serves to both authenticate the voter to the election server and the election server web site to the voter. Only someone with the correct Ki transparency could decode the password in the generated image; only something with knowledge of the transparency sent to the voter could generate a sensible password image. This process is more cumbersome, but provides substantially better security, than alternatives such as expecting a user to check a SSL certificate. In addition, we suspect from anecdotal evidence (but no scientific user studies yet) that nearly everyone will find the process of revealing a secret by holding a transparency up to an image on a monitor to be a satisfying and reassuring experience (some even find it magical!). Previous studies have analyzed how much a user needs to know in order to make rational decisions in the security of computer services, and the users showed they did not have a solid grasp on the security aspects of the system .With our system, voters do not need to understand how visual cryptography works, but are directly involved in performing the decryption in an intuitive and physical way. Our authentication scheme ensures that the voter cannot continue with the voting process without also verifying the server is legitimate.
It is important that voting is accessible to all eligible voters, and some issues must be addressed to ensure this. The first problem deals with the usability of the security of the software. The user may find it difficult to go to the extra trouble of determining what is wrong if an incorrect image is seen. An incorrect image would just come up as garbage on the computer screen, which could be improper settings of a voter's monitor. The security holds, since the image will not be seen with an incorrect layer on the server. However, the problem cannot be corrected if the user does not notify anyone, but instead just assumes something is wrong and gives up on casting a vote. The user must know what action to perform in case an image does not appear. Designing and implementing intuitive mechanisms for sizing and aligning the image and transparency is a challenging problem. Our scheme
provides a tradeoff between usability and security: larger pixels make it easier to align and view the image, but also decrease the length of the revealed password. Our approach is not a good option for voters who are visually impaired, dyslexic, or with limited motor control. Some voters may not be able to size and align the screen image, so an Internet-based voting system needs to be accessible to all voters. An alternative would need to be provided for those voters who cannot view images well enough to use our system. Traditional voting systems present a problem for people who aren't familiar with different security schemes, , lack of dexterity, etc. Remote voting system does not really address these issues directly, but we have tried to make very few assumptions about the user interface in order to leave options open for dealing with these kinds of problems. The one important requirement we impose is that every voter must have a transparency already given by election officials with him at the polling place in order to vote. This is necessary to identify the voter, to decide which ballot to present, and to generate the digital signature on the completed ballot.
The system proposed here uses visual cryptography to provide mutual authentication for voters & election servers. Internet based voting offers many benefits including low cost & increased voter participation. Remote Voting system considers security & human factors carefully. In this seminar and presentation a new framework is launched which allows voters to authenticate to voting system very safely. Visual Cryptography is the perfect solution for remote voting as it eliminates computations while decoding the information. The problem about the resolution of screen is yet to be solved. Remote voting seems to be the best application of visual cryptography.
Use Search at http://topicideas.net/search.php wisely To Get Information About Project Topic and Seminar ideas with report/source code along pdf and ppt presenaion
Active In SP
Joined: Mar 2011
07-03-2011, 10:47 AM
I need project and implimentation document for remote voting system using visual cryptography..