Biometrics is measure of biological or behavioral features which are used for identification of individuals. Most of these features are inherit and cannot be guessed or stolen.
Biometric System: It is a system that takes an individual’s physiological, behavioral or both traits as input, analyzes it and identifies the individual as legitimate or malicious user.
The biometric feature being used must be available in the database for all individuals in the community before the feature can be used for authentication. This is called enrollment.
Authentication can be in one of the following forms
- Identification: Matching an individuals features against all records to check whether his/her record is present in the database.
- Verification: To check whether the person is who he/she is claiming to be. In this case the features of the person is matched only with the features of the person they claims to be.
Types of Biometrics:
There are two broad categories of biometrics:
- Physiological Biometrics
- Behavioral Biometrics
Physical traits are measured for identification and verification in this type of biometrics. The trait should be chosen such that it is unique among the population, and resistant to changes due to illness, aging, injury, etc.
Physiological Biometric Techniques:
- Fingerprint: Fingerprints are unique for every individual. They can be measured in several ways. Minutiae-based measurement uses graphs to match ridges whereas image-based measurement finds similarities between the individuals’ fingertips image and fingerprint images present in the database. It has high level of security and used both for identification and verification. However, due to old age or diseases/injury, fingerprint may get altered. Common usage: in mobiles for verification, in offices for identification.
- Facial Recognition: Features of the face like distance between nose, mouth, ears, length of face, skin color, are used for verification and identification. Accuracy can be affected by fog, sunglasses, aging, etc.
- Iris and Retina:Patterns found in the eye are unique and can be used for both identification and recognition. Devices to analyze retina are expensive and hence it is less common. Diseases like cataract may alter iris patterns
- Voice Recognition: The pitch, voice modulation, and tone, among other things are measured. Security is medium, due to the similarity in voice of people, hence used mostly for verification. The accuracy can be hindered due to the presence of noise, or due to aging or illness.
- DNA: DNA is unique and persistent throughout lifetime. Thus security is high and can be used for both identification and verification
Traits of human behavior are measured in this case. Monitoring is required in this type of biometrics to prevent impersonation by the claimant
- Signature: Signature is one of the most commonly used biometrics. They are used to verify checks by matching the signature of the check against the signature present in the database. Signature tablets and special pens are used to compare the signatures. Duration required to write the signature can also be used to increase accuracy. Signatures are mostly used for verification.
- Keystroke Dynamics: This technique measures the behavior of a person when typing on a keyboard. Some of the characteristics take into account are:
- Typing speed.
- Frequency of errors
- Duration of key depressions
It is economical as expensive equipment is not required to take these measurements. However this trait can change with time, that is, people may start typing fast or slow over time.
Criteria for selection of Biometric:
- Universality: Each person should possess the biometric trait which is being used. For example, everyone has a face but it is not the case with GAIT biometric (for wheelchair users).
- Uniqueness: No two persons must be same in terms of the biometric trait being used i.e. everyone must be unique in terms of the biometric trait being used.
- Performance: Biometric trait must be invariant over time i.e. it shouldn’t change over time.
- Collectability: Biometric trait must be easily measurable.
- Performace: Processing of the biometric trait must be accurate and fast.
- Secure: It must be secure and can’t be copied.
- Acceptability: People should be willing to accept the biometric system.
Measurement of Accuracy:
Accuracy can be measured using two parameters as given below:
- False Rejection Rate(FRR): Number of times an individual with correct identity is unrecognized.
FRR=(Number of False Rejections/Total Number of Attempts)*100
- False Acceptance Rate(FAR): Number of times an individual with incorrect identity is recognised.
FAR=(Number of False Acceptances/Total Number of Attempts)*100
Benefits of using Biometric system over traditional authentication systems:
- Invariant: Biometric traits are invariant over time as smart cards get damaged over time but biometric traits doesn’t.
- Accountability: If there is a security breach, then biometric ensures who can be the responsible person for the breach but in traditional methods, smart cards can be stolen and used by someone else. Hence, accountable person is easily identifiable now days by using biometric.
- Easy to use: Biometric systems are easy to use.
- Convenient: User doesn’t have to remember passwords, pins and keep safe the smart cards like before.
- More secure: Biometric trait can’t be stolen or copied.
Privacy Issues Surrounding Biometrics:
Biometrics requires data of individuals like physiological and behavioral traits be stored in order for identification and verification. This may hinder their privacy, which is considered as a basic fundamental right. Also, there is fear of the stored data being used against them. Since biometric data for an individual is mostly unique, there is fear of it being used to monitor measurement of individuals. Therefore, the data must be stored securely and access to the database must be hierarchical.