Biometrics and Forensics 

Noblis’ expertise in biometric identification spans the full systems engineering life-cycle. As such, our key focus areas include the following:

Threat and Risk Assessment

Prior to implementing any security solution, an understanding of the prevailing threats and risks is absolutely essential. Otherwise, it is likely that much time and money will be spent without solving the true problem(s) at hand. In conjunction with system users, an analysis of the potential threats to system security must be performed to determine the specific security issues or weaknesses that need to be addressed. The risk associated with each threat should be assessed to determine if the solution cost is, in fact, less than that of the potential risk. Especially when assessing means to control physical and logical access, there exists a continuum of technological solutions including biometrics that can be employed singly or in combination with one another.
Back To Top

Biometric Technology Performance Assessment  

When considering the use of biometrics for a specific application, it is first necessary to develop a fundamental understanding of the technology’s performance with respect to average operating conditions and an average user population. This can be achieved through a combination of literature review and laboratory assessment. If performance appears to be satisfactory, then further laboratory testing can be conducted under more extreme environmental conditions and for a broader range of user and operational characteristics. Such laboratory testing can be expanded to scenario testing in which the candidate technology or technologies are installed in an operational-like setting and operated by individuals who are representative of the target user population. If scenario testing proves satisfactory, then the technology can be implemented as a pilot for evaluation in a limited operational environment. Performance metrics must be defined that are meaningful to both the technology and application under consideration. Generalized metrics such as False Accept Rate (FAR) and False Reject Rate (FRR) are applicable to all technologies, whereas other metrics may be meaningful only to a specific technology or implementation of that technology. In cases where such metrics may only be available internal to a system, it is possible to work with the technology manufacturer to “expose” and analyze these parameters.
Back To Top


Applications Prototyping  

Once a technology is believed well suited for a given application, the development of a rapid prototype can be useful to demonstrate operational proof-of-concept and feasibility. Such a prototype should be designed so that key performance metrics can be readily measured and tuned. This approach is especially useful as an aid to developing realistic, effective, and achievable functional and performance requirements in preparation for a system acquisition.
Back To Top

System Requirements Development  

The development of clear and precise functional and performance specifications prior to acquiring a new system is essential to program success. Human factors, subject population characteristics, operational environment, and performance tradeoffs such as between False Match and False Non-match rates are but some of the factors that must be considered. It is important that capabilities be built into the system so that the performance can be measured both at the time of system installation and throughout the system’s operation to ensure that performance requirements continue to be met. Developing meaningful performance requirements requires the integration of expertise in biometric identification with in depth understanding of the user’s application.
Back To Top

System Acquisition Support (RFP development, proposal evaluation)  

A variety of documents must be prepared in association with each government system acquisition. These documents convey such information as system concept of operation, functional and technical requirements, expected system cost, and method for evaluating proposals. The aim is to ensure that the government employs a fair and effective acquisition process and acquires a system that meets well-defined requirements. Evaluating vendor proposals requires a coordinated team of individuals who have a thorough understanding of the underlying technologies, the application requirements, and cost assessment. After a vendor has been selected, and system development has begun, it is critical that vendor progress be monitored to ensure technical, schedule, and cost requirements are met. Once the system is completed and installed, a thorough system assessment must be performed to assure that all requirements have been met.
Back To Top

System Operational Performance Assessment  

Although a biometric identification system may meet performance requirements at the time of installation, over the system’s life, there are a variety of opportunities for performance to degrade. Examples of such opportunities include biometric sensor degradation, changes in the application’s ambient environment, and changes in subject population characteristics. Especially in applications where biometrics is being used to achieve critical security, it is essential that minimum performance requirements be maintained. When system performance drops below minimum requirements, straightforward system maintenance may fix the problem (e.g., in the case of a dirty or defective sensor) or the implementation of improved technology or a retuning of the system may be required (e.g., when the template database grows exceptionally large or there are substantial changes in the characteristics of the subject population).
Back To Top

Technology Refreshment

Biometric identification technology is relatively new and is expected to mature significantly over the next five to ten years. Once biometric systems have been designed and installed, there will be opportunities to improve system performance by implementing an improvement to a specific technology (e.g., an improved fingerprint matcher) or an alternative biometric modality (e.g., iris identification replacing fingerprint identification). The requirement for adherence to specific industry standards will simplify and minimize the cost associated with such upgrades. There may be opportunities for improved performance by upgrading system components that are not biometric per se (e.g., higher speed processors or more powerful search engines).
Back To Top