Jam Resistant Communication

The ability to reliably send and receive data in a noisy / hostile environment is critical in many military and commercial applications.  Whether interference is a result of intentional jamming, environmental conditions, or congestive traffic, a robust system that can handle excessive noise and still transmit the message correctly is highly desirable.  Current approaches to the problem, such as broad spectrum, rely on shared knowledge between the sender and receiver, such as the sequence of frequencies.  Unfortunately, a shared secret approach does not scale well in extremely large networks, like a netcentric battlefield or the commercial GPS broadcast system.  These systems are highly vulnerable to intentional or unintentional jamming.

ACCR researchers have developed a new approach to encoding and decoding information that is highly resistant to noise.  The algorithm, known as the BBC algorithm, is based on a new type of coding theory called concurrent codes.  It uses standard equipment for transmission and reception and can be implemented on existing hardware.  Successful implementations of the algorithm have been demonstrated on software radios.  Current research is focused on other applications of the algorithm and investigating different implementations and possible attacks.

Signal Collection and Analysis of Radio Frequencies (SCARF)

The goal of the SCARF project is to provide a low-cost  tactical solution for situational awareness of the RF signal environment in a geographic area.  Specifically, a network of autonomous vehicles are used to collect raw RF data in the area of interest.  Collected data is fused and sent to a central processing location for analysis and characterization.  A GIS-based visualization system allows the user to interactively look at the RF environment for planning and analysis purposes.

Cyberspace Education Tools

It is critical for our graduating students to have awareness and knowledge of current tools, techniques, and threats in cyberspace.  In addition to assisting in curriculum development, ACCR researchers have developed a series of interactive classroom visualizations for teaching security concepts such as cryptography, security models, protocols, and public key infrastructure.  These tools are used by instructors and students to provide an active learning environment and enhance student understanding.

Neural Networks for Biometrics

ACCR researchers are investigating the use of convolutional neural nets to perform various biometric applications such as fingerprint recognition and bio-monitoring on the battlefield.  The unique ability of such systems to perform pattern matching and self-learning provide an interesting approach to difficult problems.

Cyberspace Applications of Unmanned Aerial Systems

The Department of Defense have made a substantial investment and commitment to unmanned systems as the future of warfare technology.  There are many interesting and unsolved problems in these systems, such as autonomous control, ad-hoc networking, data fusion, and communication.  ACCR has a UAV laboratory with a number of existing platforms for students to fly and experiment with.  Several student and faculty projects are completed each year in this important area.