November 22,2008:
The FBI is trying to get money to automate their analysis of DNA
collected from actual and suspected terrorists. These samples are collected in
Iraq and Afghanistan, as well as from suspects encountered anywhere in the
world. Currently, the JFAIDD (Joint Federal Antiterrorism Intelligence DNA
Database) contains over 15,000 analyzed samples collected, for the most part,
since September 11, 2001. Currently, manual methods only permit analysis of about four
samples a week. Automated equipment will allow over 200 samples to be analyzed
a week. There are nearly 40,000 samples
waiting for analysis, with nearly 500 urgently needed.
Other forms
of biometric evidence has, in the last decade, provided a lot of useful
evidence. The oldest biometric is your appearance, which is pretty unique. Next
came fingerprints, which were first recognized as a biometric indicator four
centuries ago, but did not become a feature of police work until 150 years ago.
That was followed by blood types and a whole bunch of stuff you could only do
with dead bodies.
But in the
past few decades, there's been a lot more. DNA, automated facial recognition,
iris patterns and many more. But all this has been accompanied by new
technologies that have made it easier to collect, store and retrieve biometric
data. That made it possible to use biometric data on the battlefield. Al Qaeda
was defeated in Iraq partly because of a huge (several hundred thousand
individuals) biometric (fingerprints and iris scan) database, collected during
raids or after arrest by U.S. troops. This took anonymity away from many
terrorists, and potential terrorists or terrorist supporters. Made it much
easier to run down the bad guys later. DNA, obviously, was only a part of this evidence,
and the most difficult to analyze.
The delay in
funding the automated DNA analysis system is largely because other forms of
biometric data was usually available. There are even more east-to-collect forms
of biometrics coming into use. These new forms of evidence are using unique
behaviors of people to identify them. The first one of these to get heavy use
was typing patterns. Actually, this one was first discovered in the 19th
century, when telegraph operators found they could recognize each other by the
pattern each used when tapping the telegraph key. This was called an operators
"chop", and was eventually applied to how people hit the keys on a
typewriter, or computer keyboard. It was eventually found that everyone had a
distinctive "chop" when using a computer keyboard, and software could
be used to recognize individuals.
The
availability of cheaper and higher resolution digital cameras made possible new
biometric identifiers, like gait analysis (we each walk with a distinctive
gait). This proved useful when using UAVs to look for elusive terrorists below.
But it could also be used just as a surveillance tool. Of course you can
deceive behavioral biometrics, but it isn't easy, and you're never sure that
your change up has fooled the software.