Biometric Security

Identifying individuals in an accurate manner is one of the main problem most of the organizations are facing today. Currently existing traditional security systems use a knowledge based or a token based mechanism to identify individuals. The crucial disadvantages of such systems are that the token or the security information might go to frauds hands and it might get lost or forgotten. Such systems are unable to identify a fraud using the security information (PIN) or the token. To avoid such fraud access, organizations tend to implement automated identification systems where biometrics identification plays a major role. Biometrics is being used to automate the identification of people by one or more of their behavioral or physical characteristics. Biometrics cannot be stolen as traditional PIN, passwords or tokens and biometrics systems deny fraud access determining whether the user is the intended authorized person.

Personal identification can be defined as the process of associating a particular individual with an identity [2]. Identification can be divided in to two operation forms; verification and recognition [2]. For a security system to become effective both user and the technology should function properly. Before a decision is taken in design, develop and implement biometric security system the technology, a detailed cost-benefit analysis and a trade off analysis should be considered [3]. Several features are considered when it comes to selecting a biometric for a certain system. A universal, unique, permanent and non distributable characteristic is best suited as a biometric [2, 4]. Number of other factors such as performance and acceptability should also be considered. A Biometric trait is more reliable, tightly bounded and less in cost than traditional security tokens.

A biometric system consists of two modules; enrollment module and identification module [2, 1]. Both modules deal with pattern recognition. During enrollment biometric sensors are used to get a digital representation of the certain bio metrics and it is further processed via a feature extractor to generate a more compact and expensive template to be stored in the database [2]. The same process for a user biometric will be done in the identification and a feature matcher is used to find a matching template in the database. If a match is found the access will be granted and if not the access to the system will be denied.

Most popular biometric technologies that have been used in access control, criminal identification, and border security are facial recognition, fingerprint recognition, hand geometry, iris recognition, retina recognition, signature recognition, and speaker recognition. A wide range of security applications have been implemented and being used. For example USA nuclear plants are using biometric hand geometry recognition to prevent authorized access to the high security area [3]. Facial recognition can be used to monitor or identify individuals even there is a large number of people on the company, clubs or in public gatherings. Some systems pick out faces in a crowd and compare them to a stored database.

The new trend in biometrics is developing multi modal biometric system which will take more than one biometric to authenticate an individual. Multi modal biometric systems are more reliable and they have a low response time. US Department of Defence Common Access Card is a multimodal biometric system [3].

Biometric security systems have various issues which are considered in ongoing researches. Biometric security will be further advanced by having multimodal biometric systems, video recognition systems as well as biometric being used for other applications such as health.


References

[1] Jain A.K. and Ross A. (2004), MultiBiometric Systems, Communications of the ACM, 47(1), 34-40

[2] Jain A.K., Hong L. and Pankanti S. (2000), Biometric Identification, Communications of the ACM, 43(2), 91-98

[3] Rhodes K.A. (2003), Information Security challenges in using biometrics, United States General Accounting Office, 1-27

[4] Tsun-Li C., The Role of Biometrics in Virtual Communities and Digital Governments
http://www.dcs.warwick.ac.uk/~ctli/papers/Chapter6.pdf