A number of identification-checking modalities have recently surfaced to address this challenge. The gold standard is multi-factor authentication, which encompasses three elements: 1) something only (presumably) the user knows, such as a PIN or password; 2) an item the user has in his/her possession, such as a flash drive, or a token that provides random authentication codes; and 3) biometrics, something physically unique to the individual.

However, the challenge begins at this point. For example, requiring a user to possess a verification tool entails limitations due to the costs of producing and distributing the necessary hardware. Every student would have to have one. Of even greater concern, such devices do not necessarily authenticate the individual, but rather verify that a person has possession of the card or device.

Which leaves biometrics. Examples include fingerprints, iris scans and facial recognition. While these offer near-absolute verification, this type of identification requires a sophisticated, expensive, hardware device to capture and interpret the biometric patterns.

As a subset of biometric physical qualities, dynamic (AKA biomechanical) biometrics offers the possibility of identification without the need of additional hardware. Within this category exist three possibilities: gesture/signature, keystroke and voice recognition.

Yet, voice recognition often results in inaccurate matching because of variations in microphone fidelity. Keystroke analysis establishes the unique patterns and dwell times of an individual while typing. However using a keyboard to enter answers to multiple-choice questions on a test, for instance, does not allow enough time to identify an individual’s unique pattern.

Which leaves gesture/signature recognition, as exemplified via the handwriting of ones name (usually initials), shapes, or series of numbers. On the face of it, one would expect that this option would still necessitate a reader. Yet, recent innovations in software engineering have created a virtual reader that users gain access to via the internet, making it instantly and universally available while still ensuring a high degree of specificity.

In independent testing by the Tolly Group ─ a leading global provider of testing and third-party validation and certification services to the Information Technology industry since 1989─one gesture recognition system, BioSig-ID™, was found to be 27 times more accurate than keystroke analysis reported in earlier evaluations. Observed confidence ratings at 99.97 percent meant that the false positive level of the BioSig-ID software was three times better than guidelines put out by National Institute of Standards and Technology.


Add your opinion to the discussion.