This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
The advancement of technology has facilitated noble technological developments such as e-voting system besides others. A number of countries such as United States and United Kingdom have been using it. However, this system has come under sharp criticism with regard to its efficiency and restoration of democracy. Major techniques in cryptography have been innovated to enhance e-voting system. This paper discusses major security considerations which are basic in developing and implementing an efficient electronic voting system. These aspects include encryption/ decryption, confidentiality, accessibility, security trends and/or attacks, and authentication of the electronic voting system. Security loopholes and necessary mitigation measures with respect to each aspect has been extensively discussed. Besides, the e-voting system (Direct Recording Electronic voting) which makes use of biometric parameters has also been compared to the traditional ballot voting system.
In the past few decades, electronic voting system has been applied by various states to conduct elections. One of its major advantages is that results are derived on real-time basis; thus, it is highly efficient besides saving costs to print ballot papers. However, with the advancement of information technology, electronic voting is faced with a lot of challenges from a security point of view as it is highly prone to cyber attacks. Hence, it is imperative to establish an efficient security framework that is holistic in nature, which combines and enhances a wide variety of security aspects such as reliability, safety, integrity, secrecy, maintainability, and stability. Such aspects come in handy to improve the security of an e-voting system. In the following paragraphs, this paper will discuss some of the major security issues related to electronic voting.
First, privacy of an e-voting system is a major aspect that should be considered at the implementation stages of an electronic voting system. This may be achieved through encryption/decryption of all electronic data within the voting system to mitigate tapping by unauthorized persons who manipulate such data maliciously. Encryption converts data into cipher text while it is being transmitted in a protocol whereas decryption converts it back into original form without alterations in a way that is understandable only to the desired recipient. In an e-voting system, it is imperative to encrypt the e-votes so that any other person (for instance, election administrators) other than the voter himself or herself may not be able to establish choice of the vote cast. Hence, the ideology of secret ballot is maintained. Cipher texts include substituting letters for numbers, rotation of numbers in the alphabet, and the scrambling voice signals by inverting sideband frequencies. Complex ciphers are applied alongside computer algorithms which rearrange data bits into digital signals and or certificates, thus masking voter identity besides mitigating auto reversing of the vote. Most commonly, encryption is achieved through using public-key and private-key encryptions to verify voters in an electronic system. Encryption/decryption is beneficial in that a decryption key is necessary to recover encrypted data, and the more sophisticated the key/algorithm is, the more difficult it is to maliciously tap electronic data being transmitted over a network protocol (Kurose & Ross, 2000).
In the public-key cryptography, the electronic data such as votes cast are transmitted using the public information, but it is only decrypted using a private key held by the recipients, although it does not fully guarantee security. An example of public-key cryptography which may be applied is Rivest Shamir Adleman (RSA). It "uses modular arithmetic, and elementary number theories to perform computations using two very large prime numbers," and it is the actual standard for encryption. RSA algorithms are, for instance Rc4 (applies random permutation), Rc5 (has variable key size), Rc6 (includes integer multiplication and 4-bit registers), and blowfish (a 64-bit cipher and applies a key length of between 32 and 448 bits) (Kurose & Ross, 2000).
Secondly, voter confidentiality is another feature of major importance in any voting system. The voting system design should be such that no one is able to determine how each person voted. Anonymity and confidentiality of electronic voters is achievable through masking every voter's identity in such a way that any attempted reverse association is impossible. Even though such approach is preferable, it however renders it difficult to attain accountability. However, this problem can be eliminated by use of public key encryption to verify at a later date that a particular vote was cast and recorded appropriately (Kohno, Stubblefield, Rubin, & Wallach, 2004). Besides, confidentiality entails that voters are capable of voting independently and any good system should prevent coercion and vote selling. Therefore, a voting system must ensure confidentiality and integrity by making certain that the votes are kept secret as well as recorded accordingly. Voter confidentiality and anonymity is important in guaranteeing the safety of the voter especially in cases where political atmosphere is highly volatile. This goes a long way in building trust in voters, thus encouraging significant turn-out. Subsequently, the voting system must also be accountable and in this regard all the logistics involved with the whole voting system must be critically monitored in such a way that voter confidentiality is not violated (Kohno et al., 2004).
There is a need to understand that the system operation is prone to a number of vulnerabilities with respect to confidentiality which is an integral aspect of a good voting system. In the existing electronic voting systems, for instance, Direct Recording Electronic System, there exists no satisfactory scheme which is complete and can guarantee the anonymity of the voters as well as consistency of votes cast compared to those tabulated. Nevertheless, it is a challenge to concurrently achieve voter confidentiality and non-subvertible monitoring in a highly integrated electronic voting system. This is because of unforeseen factors and fast software development-for instance, cyber attacks such as spoofing and social engineering. In general, confidentiality entails that no one controls or determines how the voters cast their votes and the election authorities cannot link a particular ballot to a certain voter (Cranor, 1996).
Thirdly, a good electronic voting system is supposed to be reliable as well as accessible on a real-time basis. If accessible to the diverse voting population, electronic voting can be a great improvement compared to the traditional method-printed ballot papers. However, accessibility of electronic voting system to all kinds of people regardless of age, education, profession, geographical location and or disability is a challenge whatsoever. An efficient electronic voting system should have convenient voting procedure and user-friendly execution commands. An electronic voting system which is too sophisticated goes a long way in discouraging some voters from exercising their democratic rights especially with regard to the physically challenged and uneducated persons. Hence, it is important to put into consideration the needs of both the handicapped and elderly people (Bederson, Lee, Sherman, Herrnson, & Niemi, 2003). Computer disability is another issue that can hinder accessibility of an electronic voting system. Therefore, there is a need to have a well-designed electronic voting system that is accessible and easy to use for all kinds of people regardless of their knowledge and skills level. Subsequently, the e-voting system should consider technological generation gap within the voting population eliminates malicious or accidental denial of any person qualified a voter. System availability is achieved by means of various techniques which aim at increasing hardware-fault tolerance and increasing the security of the system, but such techniques, though recommended for use, do not offer sufficient guarantee.
Integrity entails that any data within the system is highly protected (for instance, votes cast are not modifiable) and that such a system is highly reliable. The electronic voting system is supposed to be mechanically sound in order to get rid of any breakdown during the voting process. To some extent, system integrity is achievable by mitigating unauthorized software and hardware modification-tamper proof system is achievable by making use of password protection. It is common that integrity is many a times compromised by malicious programs which may include Trojan Horses and trapdoors insertion. However, this can be controllable by using read-only memories which are not modifiable. The integrity of an electronic voting system can be maintained by an embedded system code aside from enhancing the system's accountability. Despite the need for voter anonymity, techniques that permit accountability must be developed and to prevent tampering with audit-trail, read only media can be used (Neumann, 1993). Such measures comes in handy to ensure that any data fed or recorded during the voting period up to the time votes are tallied is not manipulated. Besides, the personnel involved in software development, its implementation, operations, and administration are supposed to be of high moral-uprightness.
How will the e-voting system know that the individual is who he or she is? Identification cards have always been used in voting booths, but an e-voting application that is used from home computers would require new methods that permit only genuine voters to vote. Non-trivial verification mechanism should be used to gain access by all the personnel who have been authorized to implement and administer an election process. However, using fixed passwords is inadequate and unreliable as they are easily compromised. In another case, trapdoors for maintenance and setup should not be in place as their existence can be misused to effect an operational subversion. Apart from the traditional ways of authenticating voters, there are several ways which include biometric schemes as well as classical and token approaches. Classical approaches make use of smart cards to verify voter's identity. On the other hand, a biometric system refers to instances where voters' identity is verified using specific body parts such as fingerprints (a substitute to pin codes to initiate any process), iris (makes use of retina scanner), face (makes use of infrared records), voice, and digital signatures. However, the lack of standards hampers the improving reliability of biometric system. Although such biometric schemes may be somehow manipulated, they are not much vulnerable compared to using fixed passwords. Other essential criterion that tests authenticity include Voter verification audit trials (VVAT) where a ballot paper is printed by the electronic voting system and contains computerized selections. Such print-outs are physically confirmed by the voter before being cast into the ballot box.
Finally, it is not only electronic voting that is prone to security flaws which are accelerated by the presence of free malware widely available in the Internet that targets critical elements of a system (Kurose, & Ross, 2000). In this case, major components of an electronic voting system include the "hardware, software and human factors" (Kurose, & Ross, 2000). Hardware is prone to attacks on its mechanical, electromechanical and electrical parts whereas software attacks aim to destroy the operating system, compilers, databases, rules used in the program, procedures, and or sequences (order of voting events, voting protocol, encryption techniques) (Lauer, 2004). On the other hand, security attacks on the human factors depend on the kind of personnel involved right from software development. They render the system to be unsecured by compromising honesty, security enhancement strategies, transparency, and secrecy entrusted in them. In many instances, most of the attacks are directed towards the software of an electronic voting system. Such attacks include virus, worms, and Trojan horses which are maliciously installed to manipulate votes. Others include social engineering, sniffers, and spoofing. In this regard, the Direct Recording Electronic Voting Systems is the most vulnerable to these attacks, but they can be counter-measured through various ways, such as by developing standards, legal sanctions, IT certifications, and voter verified audit trail (Lauer, 2004).
In his work, Ondrisek (2009) discusses a method known as E-voting System Security Optimization which identifies possible security flaws and optimization potential in various electronic voting systems. He further states that such common criteria is applicable in certifying IT system security and some common models include Test Process Improvement, which enhances software tests and Capability Maturity Model. E-voting System Security Optimization applies a checklist-styled configuration that facilitates creation of particular propositions to maximize safeguards of verified e-voting systems. Besides, EVSSO has a standard scheme and it is thus compatible with a wide range electronic voting systems for instance Direct Recording Electronic voting systems. However, its application in internet elections systems is highly discouraged because of an inherent insecure environment for instance, coercion, denial of service attacks, free suffrage, and lack of reliable audit-mechanisms among other risks (Ondrisek, 2009).
As earlier mentioned, new technologies aimed at improving electronic voting security are commonly available-for instance, the use of cryptography (homomorphic encryption, blind and digital signatures), antiviral programs, and firewall (Kurose & Ross, 2000). One of the major advantages claimed by proponents of electronic voting is convenience and equal participation of voters at regardless of their geographical location. However, this method of voting is questioned by its control measures to mitigate hackers accessing system information and manipulating the data there-in thus, uncertainty in terms of security and verifiability.
The introduction of electronic voting systems in nowadays elections worldwide has facilitated a paradigm shift in developing various democracies. However, compared to the traditional methods of voting in terms security, it is highly prone to cyber attacks but more efficient and such internet crimes are executed by only a few knowledgeable individuals. These attacks compromises confidentiality, accuracy, integrity, and reliability of an electronic voting system. Nevertheless, such incidences are voidable if e-voting software includes in-built antiviral programs, complex encryption/ decryption techniques, and detailed authentication processes such as biometric parameters.
- Bederson, B. B., Lee, B., Sherman, R. M., Herrnson, P.S., Niemi, R. G. (2003). Electronic voting system usability issues. In Proceedings from SIGCHI: Conference on Human Factors in Computing Systems (pp. 145-152). New York: Association for Computing Machinery.
- Cranor, L.F. (1996) Electronic voting computerized polls may save money, protect privacy. Retrieved from http://www.acm.org/crossroads/xrds2-4/voting.html
- Kohno, T., Stubblefield, A., Rubin, A. D., Wallach, D. S. (2004). Analysis of an electronic voting system. IEEE Symposium on Security and Privacy. Retrieved from http://www.cs.washington.edu/homes/yoshi/papers/eVoting/vote.pdf.
- Kurose, J., & Ross, K. (2000). Computer networking: A top down approach featuring the Internet. Massachusetts: Addison Wesley Publishing Company.
- Lauer, T. W. (2004). The risk of e-voting. Electronic Journal of e-Government, 2(3), 177-186.
- Neumann, P. G. (1993). Security criteria for electronic voting. Retrieved from http://www.csl.sri.com/users/neumann/ncs93.html.
- Ondrisek, B. (2009). E-voting system security optimization. Hawaii International Conference on System Sciences. Retrieved from http://electrobabe.files.wordpress.com/2009/01/ondrisek_e-voting-system-security-optimization_05-02-02.pdf.