ENTRY CONTROL TECHNOLOGY BIOMETRIC FIELD EVALUATIONS

Year
1995
Author(s)
D.L. Lowe - Sandia National Laboratories
Janet Steele Ahrens - Sandia National Laboratories
J.R. Rodriguez - Sandia National Laboratories
Abstract
Laboratory evaluations provide system engineers basic design information. Laboratory generated data is useful for initial comparisons of devices and in estimating the effectiveness of a device within a system. Though laboratory data is very useful, it only presents a part of the total picture. Since the performance of a device may change in transition from lab to field, system engineers also require field data. Throughout the years, Sandia National Laboratories (SNL) has performed various laboratory evaluations of entry control devices, including biometric identity verifiers. The reports which resulted from this testing have been very well received by the physical security community. This same community now requires equally informative field study data. To meet this need we have conducted a field study in an effort to develop the tools and methods which our customers can use to translate laboratory data into operational field performance. The field testing described in this report was based on the Recognition Systems Inc.'s (RSI) model ID3D HandKey biometric verifier. This device was selected because it is referenced in DOE documents such as the Guide for Implementation of the DOE Standard Badge and is the de facto biometric standard for the DOE. The ID3D HandKey is currently being used at several DOE sites such as Hanford, Rocky Flats, Pantex, Savannah River, and Idaho Nuclear Engineering Laboratory. The ID3D HandKey was laboratory tested at SNL. It performed very well during this test, exhibiting an equal error point of 0.2 percent. The goals of the field test were to identify operational characteristics and design guidelines to help system engineers translate laboratory data into field performance. A secondary goal was to develop tools which could be used by others to evaluate system effectiveness or improve the performance of their systems. Operational characteristics were determined by installing a working system and studying its operation over a five month period. Throughout this test we developed tools which could be used by others to similarly gauge system effectiveness.