The next wave in biometric security is predicted to be the use of multiple biometric authentication systems for human identification. Multi-modal biometric authentication systems take input from a single or multiple biometric sensors for measurement of two or more different biometric characteristics to ensure authentication accuracy. Multi-modal biometrics usage is being actively considered in applications involving border control, physical access control, desktop and network security. It provides supplementary information among different modalities in order to increase the recognition performance in terms of accuracy and reliability to overcome the drawbacks of a single biometric system. Research demonstrates that multi-modal biometrics as a solution to improve the accuracy and reliability of biometric systems. Identification based on multiple biometrics represents an emerging trend in high risk areas and there are many reasons to promote this – one of the most important being improving recognition effectiveness.
Behavioral biometrics or behaviometrics is a form of biometric authentication that has shown promise to address the continuous frictionless authentication problem by allowing the device to identify the user without the user doing any explicit authentication actions while providing a strong form of authentication. Behavioral biometrics identifies users based upon their behaviour rather than upon fixed physical characteristic (such as a fingerprint). Behavioral biometrics learn patterns in user behaviour in order to build a user identification model and authenticates the user based on whether their behaviour conforms to the recorded model of the user behaviour. Accuracies ranging into the high ninety percentile are common using behavioral biometrics with room for improvement.
Reasons to Adopt Multimodal Biometrics
- Accuracy and Reliable Recognition: A multi-modal biometric system permits a higher level of assurance for an accurate match in verification as well as identification modes. As multi-modal biometric systems utilize multiple biometric traits, each single trait can offer additional evidence about the authenticity of any identity claim. For example, the patterns of movements (gaits) of two individuals of the same family or coincidentally of two different persons could be similar. In such situations, a unimodal biometric system based only on gait pattern analysis might lead to a false recognition. If the same biometric system additionally includes fingerprint matching or finger vein matching, the system would certainly result in increased recognition rate, as it is nearly impossible that two different individuals have same gait as well as fingerprint/finger vein pattern.
- A fortified level of security: Rather than having just one layer of security, you can now have multiple layers. Not only will there be a fail-safe mechanism in place, but if a perpetrator successfully breaks through one line of defence, they will most likely be caught deeper in. Also depending on the security requirements – one could choose the level of security. For example, in an extremely high security site, some use up to three biometric identifiers and for a lower security site, you could possibly require one or two credentials. If one of the identifiers fails for any unknown reason, your system can still utilize another one or two of them in order to provide the accurate identification of a person. This could potentially reduce the probability of admitting an imposter.
- Vulnerability: In biometrics, there is always the threat that a system could be spoofed. For example, the potential threats due to fake or artificial fingers were evaluated by a team of researchers, whose experiments showed that artificial fingers cloned with plastic moulds could possibly enroll in the 11 tested fingerprint systems and were being accepted in the verification procedures with the probability of 68% -100%, depending on the system. By having extra layers of biometrics, this issue has become over time, much less prevalent.
- User Acceptance: As multimodal biometric systems are more accurate, reliable, have larger security options, and have the ability to avoid spoofing attacks, these systems are more widely accepted in many countries that cover large to larger deployments. Biometric deployments that encompass large scale population databases are turning to multimodal systems. However, in deployments where security and accuracy are paramount, no matter how small, multimodal systems have become ubiquitous.
Multi-modal biometrics and industrial deployments
Some examples of deployment of multimodal biometrics technology in the industry:
- Due to the increase in border problems worldwide, over 80+ countries have implemented multimodal biometric technology to enhance border security. Countries like Hong Kong, Japan, Australia, and the USA have successfully deployed multimodal biometric identification systems at immigration and border security.
- Another example is the Bahrain based Mass Contracting Co. – a Manufacturing and construction company that wanted to take attendance of their construction workers and chose to deploy a multimodal solution with both a fingerprint and finger vein modality. They made that decision after considering the nature of work performed by their construction workers. Construction workers are expected to have dry, cut or damaged fingers. Having only a unimodal fingerprint based solution was not ideal to capture the majority of individual biometric credentials.
For mobile commerce and payment where there is a need for secure always on continuous authentication and fraud detection solution that is frictionless and balances convenience, security, and privacy – behavioral biometrics have the advantage that they can work implicitly with other static biometric options such as fingerprint or iris scan. This allows for very rapid authentication decisions.