|
|
Biometrics
and the Border Management Challenge
Testimony
of
Dennis
Carlton
Director of Washington Operations
International Biometric Group, LLC
To
the
House
Select Committee on Homeland Security
Subcommittee on Infrastructure and Border Security
"Integrity
and Security at the Borders: The US VISIT Program"
January
28, 2004
My name is Dennis
Carlton and I am the Director of Washington Operations for
International Biometric Group of New York City. On behalf
of our company, I'd like to thank the committee for the
opportunity to talk to you about the technology called
biometrics and some of the likely the issues associated with
introducing biometrics into the border management system.
Let
me begin with a brief description of International Biometric
Group so that you better understand who we are and our unique
position in the world of biometrics. International
Biometric Group, or IBG, provides independent consulting
services to government and private industry customers interested
in implementing biometric technologies. Our organization
focuses on three primary functions: (1) evaluating and reporting
on biometric products and vendors, as well as the markets in
which they compete, (2) advising clients on how to implement
biometric systems, and (3) integrating a wide range of biometric
hardware and software to meet the security needs of our
customers. We take a practical, hands-on approach toward
biometrics. We have conducted extensive comparative
performance testing of more than fifty different biometric
solutions so that we know how they're likely to perform in the
real world. IBG holds to a strict vendor-neutral policy,
which enables us to maintain close relationships with biometrics
vendors while ensuring that our clients receive accurate and
independent advice on which biometric systems can best meet
their needs.
I'd like to
take a moment to review some of the basics of biometrics.
A technical definition of biometrics is the automated
measurement of behavioral or physiological characteristics of a
human being to determine or authenticate their identity.
In other words, it's the use of computers to confirm who a
person is by matching a behavior or a permanent physical
characteristic with similar records in a database. Biometrics
alone can't determine an individual's identity but they can
effectively distinguish one person from another. There is a wide
range of products in the market that can acquire and match a
person's biometric data to perform a quick and accurate
identification. With respect to border management, the U.S. has
focused its attention on fingerprint matching and facial
recognition biometrics, although other biometrics such as iris
recognition, hand geometry, and speaker authentication
technologies are also being assessed.
One year ago, IBG
delivered a report to the White House Office of Science and
Technology Policy entitled "Use of Biometric Technologies in
the United States Visa Issuance and Border Entry/Exit
Systems". I was the principal investigator and author of this
report, a summary of which has been included in the material
provided to committee members. In conducting research for this
study we visited several U.S. consulates around the world as
well as American sea, air and land ports of entry. The OSTP
sought a no-holds-barred look at the practical challenges of
implementing biometrics in the field both at consulates and
ports of entry - I like to think we accomplished that goal.
From the OSTP research and our subsequent participation in
several ongoing initiatives involving biometrics and
international travel security, IBG has gained significant
insight into the integration, performance, and workflow
challenges associated with implementing biometrics within US
VISIT and our border management system.
IBG's report to
the OSTP highlighted several issues related to integrating
biometrics within US VISIT worthy of reemphasis before this
committee:
- Biometrics
should be implemented in a manner that augments rather than
replaces existing border management IT systems. The fact
that an individual matches the biometric associated with a
travel document does not ensure that the individual
qualifies for admission to the United States. Biometrics
alone cannot replace the professional judgment of
experienced border management personnel.
- Since
the current generation of biometric technologies is not 100%
accurate, a seamless exception handling process must be
incorporated throughout the design of the system.
- The
system also must be designed with an eye toward continuing
technology refreshment. The lifecycle of biometric products
turns over at least as fast as other IT components - US
VISIT should be designed with seamless transitions to newer,
more accurate solutions in mind.
- The
government must invest in continuing research and
development into improving biometric products. The
centennial of the Wright Brothers first powered flight
serves as a reminder that significant innovations may come
from unlikely sources. In the past few years there has been
an explosion of new biometric technologies being introduced
into the marketplace, many of which warrant the nurture of
the federal government. In addition to financial support,
the federal government may need to approve regulatory and
legislative changes to authorize the development of
databases that can be used to test the effectiveness of new
biometric solutions.
The success of
any biometric solution depends in great measure on its
stakeholders establishing realistic performance expectations for
the system; given its unprecedented scale and visibility, this
will be especially true of the US VISIT system. Among the key
performance considerations are:
- Stakeholders
need to have practical expectations as to the performance of
biometric technologies. The current generation of biometric
systems is not 100% accurate but biometrics don't need to
be perfect in order to enhance border security. The mere
presence of a device that can positively link an individual
with the documentation they carry will serve as a deterrent
to many impostors. Border inspection personnel use their
professional judgment to resolve exception situations every
day; biometrics problems can be resolved in much the same
manner as any other identification document discrepancy.
- The
system design must incorporate a comprehensive security and
privacy architecture. Good security and privacy practices
are not antithetical and can both be accommodated in US
VISIT. Biometrics themselves are privacy neutral - it's
the way they are employed, and the protections put in place
to limit misuse, that make biometrics either
privacy-invasive or privacy-protective. What is essential is
that individuals are fully informed on how their data is
collected, used, shared, and secured. For more information
about biometrics and privacy I commend to you an IBG-sponsored
website dedicated exclusively to the subject, www.BioPrivacy.org.
- Reaching
a consensus with our international partners on privacy
policy will be difficult because of significant differences
in our privacy expectations. In general, while Americans
often don't hesitate to provide personal data in exchange
for commercial benefits but frequently oppose sharing such
data with government, their counterparts in Europe and Asia
view cooperation with their government as a duty of citizens
in a civil society but don't feel similarly compelled to
provide personal data to commercial concerns. For the time
being, some parties have staked out extreme positions.
Compromise will mostly likely be achieved when views
converge toward a standard that defines a minimal exchange
of a traveler's personal or biometric data to effect
efficient commerce between governments.
The advent of US
VISIT and biometric technologies will certainly alter the
primary and secondary inspections processes at U.S. ports of
entry but these changes need not result in delay and
inefficiency. Some of the measures that can be taken to reduce
the impacts caused by these changes include:
- U.S.
border management solutions must be designed to accommodate
multiple forms of biometric technologies. Although the
International Civil Aviation Organization has specified that
facial recognition is the universal biometric to secure
machine-readable travel documentation, the U.S. will
continue to leverage its investment in fingerprint databases
to identify travelers who might pose a security threat. A
universal biometric solution is not necessary in order to
achieve a secure border management solution so long as
countries agree to provide one another with the software
necessary to decode and match the specific biometric data
associated with a travel document - this approach would
allow Visa Waiver Program participating countries to confirm
the identity of one another's citizens. A travel document
that is secured by multiple forms of biometric technologies
would significantly complicate the job of a forger or
impostor. Eventually the U.S. will need to employ biometrics
to secure the travel documents it issues its own citizens or
risk having a U.S. passport become the document of choice
for fraudsters or terrorists seeking to avoid being exposed
by biometric identification.
- The
challenge of implementing biometric identification at land
ports of entry is daunting but not insurmountable. IBG
believes that portable fingerprint reading devices can be
employed to capture images of the index fingers of all
passengers in a vehicle in order to authenticate them with
the travel documentation they carry and to check against
watch lists of undesirable individuals. The capture of
fingerprints must take place ‘upstream' from the primary
inspection station so that a biometric search can be
conducted before the vehicle reaches the primary inspection
position. In this way, the biometric search does not impact
the overall primary inspection time and the primary
inspector is not distracted from conducting a thorough
assessment of the vehicle, its occupants, and its contents.
- At
exit points or other U.S. government service centers such as
post offices, the government should provide a self-service
kiosk that will allow exiting U.S. citizens a means for
self-enrollment of their travel-related documentation and
biometrics. In exchange for taking the time to scan their
identity documentation (e.g., driver's license, passport,
etc.) and providing a biometric sample, the U.S. citizen
could use a ‘Blue Lane' that would offer expedited
processing when returning. The success of solutions like the
SENTRI system on the U.S.-Mexico border and the
U.S.-Canadian cooperative program called NEXUS shows that
both U.S. and foreign citizens are willing to provide
personal and biometric data to the government in exchange
for the very tangible benefit of expedited border crossing.
If they become widely used, these solutions could help make
a dent in the rise in identity theft crimes by making it
much more difficult for an identity thief to travel
internationally on a stolen U.S. identity document.
- We
should expect that most problems associated with
biometrically secured travel documents would be of an
innocent or inadvertent nature rather than a fraud attempt.
It won't be a sufficient solution to just turn away at an
airline counter or border post a traveler who has a problem
matching a biometric sample with a travel document. The U.S.
should provide travelers with a real-time problem resolution
solution - a phone number or email address where they can
immediately reach someone in an ombudsman-like role who can
begin the process of resolving their travel documentation
problems.
In summary,
biometrics will play an increasingly important role in enhancing
the integrity of U.S. border management systems. With clear
guidelines and careful compliance with the rules of how, when,
and where biometrics will be collected and employed, these tools
can improve border security while at the same time protect the
privacy and dignity of the legitimate traveler.
I look forward to
responding to the committee's questions.
|