The Characteristics and Attributes of Biometrics

Introduction
Sareen (2014) describes biometrics as the technology or the science that determines and assesses biological data. Biometrics is measurable behavioral and physiological characteristics applicable in the verification process of people’s identity (Sareen, 2014). In layman terms biometrics may be understood as a science that involves measuring behavioral and physical characteristics unique to an individual (Xiao, 2007). Biometrics verifies that any person is who they claim to be. The term biometrics refers to life measurements although the terms have been associated with the unique physiological qualities used in identifying a person. Biometric applications are often associated with security. However, the identification of the unique human characteristics scientifically has broader meaning and relevance bearing in mind that computer interface has become increasingly natural (Sareen, 2014).
Jaiswal (2011) describes biometric as any robust, measurable, physical, and distinctive personal traits or characteristic of individuals usable in their identification or verification of their claimed identities. There are two categories of biometrics as described by Sareen (2014) i.e. physical and behavioral biometrics. Behavioral biometrics can be defined as a biometric that evaluates naturally acquired characteristics over some time. Behavioral biometrics is used for verification purposes (Jaiswal, 2011). Physical biometric is used for the measurement of intrinsic physical traits of individuals and can be employed for verification and identification (Xiao, 2007).
How Biometrics Work
Notwithstanding the scheme of use of any biometric, all of them have to be processed in the same way. The biometric capturing process involves capturing, processing, and finally comparison. In capturing, the biometric scheme is utilized in the capturing of physiological and behavioral features (Jaiswal, 2011). In processing, the captured behavioral or physical features are processed in order to derive the unique identifiers that correspond to the individual person alone. In comparison, the individual will then be enrolled into the system as the user who is authorized. In the comparison process, the captured image will be checked against the current unique identifying elements. The newly captured elements are the detailed used for authorizing the user. Upon completion of the entire journey of biometric capturing, processing, comparison, the stored data is then applied for future comparison (Jaiswal, 2011).
Having knowledge of the person with whom a conversation is being undertaken is fundamental in the process of human interaction. Computers are expected to integrate these capabilities in future. There are several biometric traits that have been cultivated and employed for the authentication of the identity of people. The concept is to use people’s special characteristics for their identification. By the employment of the special characteristics, it means application of features such as their signature, fingerprint, iris, and face (Sareen, 2014).
Characteristics of Biometric
There are several biometric characteristics that could be captured during the first processing phase. In order to make automated biometric capture and automated biometric comparison with data previously stored possible the stored biometric characteristics must satisfy the following conditions (Sareen, 2014).
· Universal: Every individual has to possess the said attribute. This attribute has to be one that can rarely become lost to disease or accident and has to be universal to all.
· Invariance of characteristics: The biometrics must be constant for an extended duration of time. These biometric traits must never be prone to significant difference due to chronic disease or gaining (Sareen, 2014).
· Measurable: The characteristics must be appropriate for capturing without any delays and the attribute data should be gathered passively.
· Singularity: Each attribute expression has to be uniquely identifiable to the individual alone. The biometric characteristics have to be significantly unique in order to make it possible to clearly distinguish a single person from all others. Weight, height, eye color, hair and body figure are some of the attributes capable of uniquely identifying someone assuming they present a precise measure. They however don not provide sufficient identifiers for differentiation in order for them to be classified as useful traits other than for the purpose of categorizing (Sareen, 2014).
· Acceptance: The method of capturing the biometric data should be acceptable to most or all the people in a population. Invasive technologies are secluded from biometrics. The exempted technologies are those that might require some human body part to be extracted or technologies that might impair human body.
· Reducibility: The data that has been captured ought to be reducible into some file which can be easily handled.
· Tamper resistance and reliability: The attributes of the biometric should not allow for masking or manipulation. Biometric capturing process must be secure enough to become highly reliable and enable reproducibility (Sareen, 2014).
· Privacy: Biometric capturing process must not violate people’s privacy.
· Comparability: The biometric processes should have the ability to minimize the characteristics into some state which allows digital comparison to other biometric traits. The lesser the probability of matching with other biometric data the more the biometric becomes authoritative in identification.
· Inimitability: The biometric attributes for a single individual should not be replicable by any means. If the attribute is rarely reproducible, then the biometric becomes more authoritative in the midst of several biometric technologies. This therefore means that the biometric attributes satisfy the criteria qualifying it as a unique identifier (Sareen, 2014).
There are various unique identifiers that comply with all the characteristics that make physical and behavioral characteristics unique to an individual. Some of these features include the fingerprint, hand geometry, facial features, iris, voice, vein patterns, retina, DNA, palm print, signature, odor, shape, ear, and keystroke dynamics etc. (Dorizzi, 2013).
As previously stated by Jaiswal (2011) a biometric has to be robust, measurable, and distinctive in order for it to uniquely identify and verify a person. On measurability Jaiswal (2011) states that the trait or characteristic should be presented easily to some sensor after it is converted into some digital and quantifiable format. This makes it possible for the automated process of matching to happen in a few seconds. On robustness of the biometric process, a robust biometric measures the extent that personal traits or physical characteristic is vulnerable to considerable changes over some time. This nature of change might happen due to the impact of exposure to gaining, injury or chemical elements (Jaiswal, Bhadauria & Jadon, 2011).
A biometric that is highly robust cannot be subject to significant change with time. Lower robustness degree is an indication that the biometric could be altered to some considerable level over time. For instance, the pattern of the iris goes through minimal change over the lifetime of a person hence making them quite robust compared to a person’s voice (Jaiswal, 2011). The distinctiveness of a biometric is the measure showing the differences or variations in the pattern of a biometric in general population. Higher levels of distinctiveness are taken to mean that the biometric qualifies as the unique identifier. Lower distinctiveness levels are an indication that the biometric pattern is more frequent among a population hence disqualifying them from becoming unique identifiers (Jaiswal, 2011).
Biometric application purpose is to determine the extent of the distinctiveness and robustness required. Identification is significantly different from verification. Identification can be described as the querying from a device and an attempt and solving the query. When a biometric is used for the purpose of identification of people, the device used reads the sample and compares it against every other biometric template in that database. This comparison of biometric data is referred to as 1: N (one to many) (Jaiswal, 2011). The device might find matching biometric hence identifying the person to whom the biometric belongs or find some matching biometric that does not identify that person. Verification is biometrics refers to when a device queries and tries to solve the query. Using biometrics for verification of identity of individuals requires the device to have some entry data from the person. For instance the user will claim their identity through some token, username, or password or a combination of all of them.
The device will also require some biometric sample. The sample is them compared against the template that defines the user from the database. The template is pointed to by the username, token or password. This verification process is referred to as 1:1 (one to one) (Jaiswal, 2011). The device might find or fail to find the matching data from the database. Generally there are three approaches used in the recognition of an individual for the purpose of security otherwise referred to as authentication (Jaiswal, Bhadauria & Jadon, 2011). The approaches are listed from the least secure and least convenient approach to the most secure and most convenient approaches. The first and least secure approach uses something on the individual like the token, key, or card. The 2nd approach utilizes something that the individual is aware of like a pin or password. The third verification process uses something the individual is, and that is a biometric (Jaiswal, Bhadauria & Jadon, 2011).
Conclusion
A combination of the three recognition processes will add to the security layers and requiring the three recognition processes offers better security standards (Jain, Ross & Prabhakar, 2004). Authentication of biometrics is an automated identification method or the verification of living person’s identity in the present based on their personal traits or physical characteristics. Living person in real time is a phrase used in the differentiation of biometric authentication and forensics. Forensics doesn’t require identification of living individuals in real time unlike biometrics (Jaiswal, 2011). The biometrics technology is highly applicable in immigration, healthcare, financial services, social services, control of physical access, network access, telecommunication, law enforcement etc. Biometrics can be used a deterrent to criminal activities and terrorism as well.




References
Dorizzi, B. (2013). Introduction to Biometrics. Signal and Image Processing for Biometrics, 1– 13. https://doi.org/10.1002/9781118561911.ch1
Jain, A. K., Ross, A., & Prabhakar, S. (2004). An Introduction to Biometric Recognition. IEEE Transactions on Circuits and Systems for Video Technology, 14(1), 4–20. https://doi.org/10.1109/TCSVT.2003.818349
Jaiswal, S. (2011). BIOMETRIC: CASE STUDY. Journal of Global Research in Computer Science. Volume 2, No. 10. http://www.rroij.com/open-access/biometric-case-study-19- 49.pdf
Jaiswal, S., Bhadauria, D. S. S., & Jadon, D. R. S. (2011). Biometric: Case Study. Journal of Global Research in Computer Science, 2(10), 19–48.
Sareen, P. (2014). Biometrics – Introduction, Characteristics, Basic technique, its Types and Various Performance Measures. International Journal of Emerging Research in Management &Technology, (34), 2278–9359.
Xiao, Q. (2007). Biometrics-technology, application, challenge, and computational intelligence solutions. IEEE Computational Intelligence Magazine, 2(2). https://doi.org/10.1109/MCI.2007.353415