Biometric systems are systems that incorporate some kind of biometric scanning. That is to say, they scan and detect unique physical or behavioral characteristics for the purpose of identifying an individual. Biometric systems are almost always combined with some other systems. The biometric system usually acts as the gatekeeper in the overall system. The job of biometric systems is usually access control to protect some physical or virtual space. You can use biometric systems to protect buildings, individual rooms, or access to computer systems.
Biometric systems may scan for one biometric or a combination of biometric indicators. Some examples of biometrics include fingerprints, retinal scans, voice, and gait recognition (identifying someone by the way they walk).
False acceptance rates are a problem in biometric systems – that is to say, if a system allows access where access should not have been granted. This could be someone innocently gaining access to something they should not be allowed access or it could mean someone gaining access with malicious intent to steal, spy, or damage. The best biometric systems will have very low false acceptance rates. Of course, it depends on the sophistication of the hardware and software as well as the number of biometrics that the biometric system checks.
The opposite problem of the false acceptance is the false rejection. This is where a valid user tries to gain access but is rejected. Again, the false rejection rate in biometric systems largely depends on the sophistication of the system.
A system that has poor quality scanners or inferior software is more likely to provide a false rejection. Ironically, a false rejection is more likely in a system with more biometric requirements because there are more things that could go wrong. For example, a valid user with a cold wants to gain access to a biometric system with a fingerprint scanner and voice recognition. They scan their fingerprint with no problem, but their voice has changed because of the cold. The voice recognition software cannot match their voice as it is now with their voice recorded in the database and therefore keeps them from access.