Ad-Hoc Wireless Networking Technology For Battlefield Environments

The project for the U.S. Army is one of 31 approved last week by DoD's Multidisciplinary Research Initiative (MURI) program, a $146 million, five-year effort targeting topics of "exceptional opportunity" for DoD technologies and applications. "MURI topics have very aggressive technical goals for objectives that cannot be accomplished with existing technologies," said James Zeidler, principal investigator on the project and a research scientist and senior lecturer in the Electrical and Computer Engineering department of UCSD's Jacobs School of Engineering. "We are developing some fundamentally new concepts and approaches, and it will be a very exciting project."

Ad-hoc networks form when communications peers find each other without the benefit of centralized network control systems. Deployment is vastly more complicated in fluid tactical situations where special challenges emerge, and where the price of dropped communications can mean loss of life and mission failure. Among the difficulties: peers must stay in touch while having to move around, maintain stealth, and avoid enemy jamming and eaves-dropping attempts. The network must sustain itself when communications partners go out of range or are damaged or destroyed. Noted Zeidler: "The technology could also be useful to firefighters, police, and other responders to fires, earthquakes, terrorist attacks or other emergencies that have knocked out pre-existing communications infrastructure."

Equipment that would be used in a tactical ad-hoc network ranges from mobile radios mounted in backpacks, laptops, and hand-held computers to antennae mounted on vehicles such as tanks and Humvees that come and go, and even airborne relays to route data to and from command and control bases and headquarters.

One target of innovation for the project is the network protocol suite, a reference to the many programs used to manage communications. These typically are grouped into discrete steps or layers, an arrangement that robs a network of an ability to adapt to changing conditions. One goal of the project is to set up a cross-layer algorithm that will enable the different layers to join together in decisions, for example to decide to rotate an antenna in response to lagging signal strength, or to rapidly switch partners in response to movement.

The team will also explore the use of new antenna technology, coding, and error-correction systems. One focus will be multiple-input/multiple-output (MIMO) devices endowed with multiple antennas. This is new technology that shows great potential for enabling communications in less than optimum circumstances. In a recent study by Brian Banister, one of Zeidler's graduate student researchers, a ten-fold boost in channel capacity was achieved using an eight-fold asymmetric MIMO channel compared to a one-way channel, given the same signal-to-noise ratios on each channel. MIMO devices format signals using space-time coding, where the receiver analyzes positional differences between pieces of a signal and the timing of the signal's arrival when decoding messages.

The funding will support the work of faculty and graduate students on the UCSD campus, including at the Center for Wireless Communications (CWC) and the California Institute of Telecommunications and Information Technology [Cal-(IT)²]. Four Jacobs School electrical and computer engineering faculty will work with Zeidler on the project: Rene Cruz, Larry Milstein, John Proakis, and Bhaskar Rao. Research faculty members at other locations include UC Irvine's Hamid Jarfarkani; J.J. Garcia-Luna-Aceves at UC Santa Cruz; Yingbo Hua and Srikanth Krishnamurthy at UC Riverside; as well as A. Lee Swindlehurst and Michael Jensen at Brigham Young University. Additional funding from the Canadian Defense Research establishment will underwrite research by professor Simon Haykin at McMaster University in Hamilton in Ontario, Canada. Michele Zorzi of the University of Padua, Italy has collaborated with Professors Cruz and Krishnamurthy on numerous publications in this technical area, and will also participate in this project as a visiting professor at UCSD.