Biometric technologies are often used to confirm or identify individuals in situations where security is paramount. A biometric can be physical, measuring various characteristics of the human body. It can also be behavioral, such as the gait of an individual or speech patterns. The impact of these types of technologies on security is obvious. Sensor network technology is another important element of information security. These types of networks can be deployed in battlefields where they can track enemy movement or over large cities where they can monitor air quality for the presence of windborne contaminants. Software-defined radio is also a technology that helps ensure the security and safety of individuals, providing maximum flexibility in wireless communications to match the needs of an uncertain situation or environment. All of these technologies are based on the continued capability to store and manipulate more data faster and more robustly than ever.
A team led by J. Nicholas Laneman, associate professor in the Department of Electrical Engineering, and graduate students Brian Dunn and Michael Dickens has developed a software-defined radio technology that enables wireless devices to communicate using almost any wireless protocol simply by running a different software program. This has long been a problem with police officers and other first responders.
Because of the nature and scope of different situations, emergency personnel rarely have the opportunity to plan and setup their radios to address specific needs prior to an event. Even radios that conform to national standards for interoperability don’t always talk with each other consistently. Software-defined radio technology immediately benefits police, fire, and other emergency management departments that have struggled for decades with incompatible communication devices.
An outgrowth of this research is a company called RFware, which has recently received contracts from the U.S. Navy and Indiana’s 21st Century Fund in order to continue developing this technology for commercialization.
Because of the rich patterns in the iris, it is unique to each individual and considered one of the most ideal parts of the body for biometric identification. Iris recognition uses near infrared illumination and special camera technology to create digital images of the intricate structure of the iris. These images are then used in security applications such as frequent traveler programs to clear immigration services.
A leader in iris biometric research, the Notre Dame biometrics group in the Department of Computer Science and Engineering has studied a range of basic phenomena, including the problems presented in iris recognition technologies by pupil dilation. While an iris texture was originally thought to be an “unchanging” biometric marker, results obtained by the Notre Dame research group show that pupil dilation and age affect images of the iris. Findings by engineering faculty and students have led to changes in the ISO standard for iris template data so that it now includes a value for estimated pupil dilation.
The Notre Dame group has also studied how the accuracy of iris matching changes over time, between enrollment (the initial capture of the image) and recognition (when a “fresh” image is acquired for comparison to the enrolled image). Their work indentified a noticeable degradation in matching quality after four years, suggesting that iris images should be re-captured (a subject re-enrolled in iris biometrics databases) at specific intervals.