Home' RTCA Documents for Review : DO-230H FRAC Contents 53
©2017 RTCA, Inc.
The time that it takes to process and match a biometric sample will depend on a number of variables.
It is faster to perform a verification (1:1 comparison) biometric match than to search an entire
population of registered users to perform a biometric match (1:N comparison).
For 1:1 biometric matching, the throughput is independent of the size of the enrolled population,
because only one comparison is made against a single known record based on a “pointer” or unique
identifier read from a badge or entered by the user. However, for a 1:N implementation, the presented
biometric sample is compared to all known biometric records in the system and there is no requirement
to provide a pointer to a specific record. This implies a system where the biometric (e.g., iris
recognition) compares the presented biometric sample with the entire database every time the user
seeks entry. In such a system there would be no requirement to present a badge or enter a PIN to claim
a specific identity. In a 1:N implementation, throughput can be enhanced by eliminating the need to
present a badge or enter a PIN. This is partially offset by the additional search time. However,
matching accuracy becomes significantly more important in a 1:N biometric system since the number
of match comparisons can be large. Iris recognition is considered by some to be the best biometric
modality for 1:N access transactions since the search speed is extremely fast and the matching accuracy
is very high.
It is recommended that biometrics for access control with large populations (in the thousands) be
deployed as a 1:1 matching system unless there is confidence that the FAR of the technology is in the
range of one in one million or better and the search time for matching can be performed in one second or
The outdoor environment is particularly challenging for some biometric technologies. Care should be
taken to consider the outdoor effects on biometric technology related to changes in lighting,
condensation, low humidity-based electrostatic discharge, and airborne contaminants that may impact
biometric sample acquisition and/or matching performance. The locations where biometric readers
will be installed can make a big difference regarding final product selection. For example, any reader
that is placed outdoors will require some protections against the weather, as well as periodically
clearing foliage, providing additional lighting or protection from direct sunlight, etc. Dirt, dust, grease,
and airborne contaminants, can all have an impact on the type of biometric sensor that is used. Airport
operators should consult with system integrators or potential vendors well in advance of purchase and
installation to ensure that the selected modality can be implemented as envisioned.
Another environmental consideration is the use of intrinsically safe biometric readers that will be used in
areas that have explosive vapors present in the atmosphere (e.g., fuel loading areas). It should be noted that
this consideration is not unique to biometrics and impacts all electrical equipment used in such areas. The
reader devices should be certified for use in such environments and shielded from sparks that might ignite
the vapors. There are now a number of IEC standards, in particular the 60079 Series, which covers the use
of equipment in explosive atmospheres. These standards are recognized in most countries around the world
and have made it possible to select appropriate electrical equipment for such environments.
Using a biometric device should be fast, efficient, and most of all, easy and intuitive with the fewer steps
the better. System designers should consider how the hardware, software and design of the system impact
the user experience. Adopting a user-centric view of the biometric process can also help to improve the
performance and effectiveness of a system. For example, the biometric sensor should be readily accessible
Links Archive Navigation Previous Page Next Page