By Marlys Amundson
Computer science professors Deborah Estrin and Mario Gerla and electrical engineering professor Mani Srivastava are part of a new International Technology Alliance (ITA) in Network and Information Sciences. The ITA brings together researchers from academia and industry in the United States and the United Kingdom to address problems in network theory, secure systems, sensor information processing and delivery, and distributed coalition planning and decision making.
The alliance’s research will support military operations, which depend on the ability of forces to quickly gather, interpret, and share battlefield information to coordinate actions.
“We’re looking at issues of sensor data integrity,” noted Srivastava. “We want to be certain that the information received from the sensors is meaningful and correct, since there are many ways it could be compromised. The sensors, embedded in the physical world, are unattended and vulnerable.”
Srivastava and Estrin will be examining the various problems that might affect sensor integrity – from benign causes to deliberate attacks on the system, such as acoustic interference for a sound sensor or spurious inputs to a chemical sensor.
“For this project, we’re working primarily with acoustic, magnetic, and chemical sensors and their vulnerabilities,” explained Srivastava. “For securing against spurious data input to the sensors, we can’t rely on encryption alone, so we’re looking at notions of security based on physical properties and statistical models.”
The interdisciplinary team is looking at four key areas: the ability to tell when sensor data has been compromised, a way to make the best use of available data, the ability to determine what went wrong and why, and ways to rectify the problem.
To detect when sensor information may not be accurate, the UCLA researchers are using statistical analysis similar to that used by banks to detect fraud. Estrin and Srivastava will also partner on developing technologies that are fault tolerant so that even if some data is corrupt, the rest can be used to generate accurate and useful reports.
“We also want to be able to determine what caused the problem and why it occurred,” said Srivastava. “And we’ll be looking at ways to correct any problems, either by replacing a sensor or reconfiguring it, or restructuring the network to work around a faulty sensor.”
At the NSF Center for Embedded Networked Sensing, headquartered in the UCLA Henry Samueli School of Engineering and Applied Science, Estrin and her colleagues have been developing wireless sensor systems and the application of revolutionary sensing technology to critical scientific and societal applications. Their more recent work on applications for urban sensing involves many of the same privacy and security issues the researchers will be looking at through the ITA.
“It’s a set of technologies that will play a role in military situations, as well as civilian,” Srivastava explained. “We’re likely to have overlap in the types of sensors we’ll be working with for sensing of urban spaces, health monitoring, use in civilian security areas, and localization of vehicles in tunnels and canyons where GPS doesn’t work.”
“This is the first major collaboration between the U.S. and the UK at this scale on a military project,” said Gerla. “The funding agencies are very interested in seeing where it will lead.”
Gerla and his colleagues will be looking at mobile networks to support communication between soldiers, tanks, airplanes, and other units, which can be subject to disruption or jamming.
“We’re designing protocols that are robust to support the broadcasting of messages across the nodes,” Gerla explained. “For instance, how quickly and efficiently can an alert be sent to the troops. Is a message stored and then passed along when two units come into contact? Or should we use unmanned aerial vehicles to serve as communication points?”
Mobile wireless networks must address efficiency issues as both radio frequency spectrum and energy are extremely limited. One technique the researchers are considering to better utilize available spectrum is “spectrum scavenging” or searching for and employing spectrum segments not currently used.
“By using new technologies – for instance, cognitive radios, which automatically adjust to maintain communications – we can set up the network to make better use of all available resources,” said Gerla. “Such a system would also provide stronger protection from enemy jamming, since it can adjust on the fly without human intervention.”
Overall, a new generation of network protocols are being designed that must take into account both the unpredictable nature of radio propagation and a variety of enemy threats, such as spectrum jamming and radio or robot capture.
“A stimulating aspect in this project is that the new protocols will be built on solid theoretic foundation before we move to implementation and testing.” Gerla added, “And eventually, the work we are doing will be transitioned to commercial networks, such as vehicular networks in urban environments.”
The 24-member consortium will address both fundamental research and technology transition. The program could last 10 years with up to $135.8 million in research funding from the UK Ministry of Defence, U.S. Department of Defense, and some consortium members.
The ITA includes IBM, BBNET Solutions, The Boeing Company, Honeywell, Klein Associates, LogicaCMG, Roke Manor Research, Systems Engineering and Assessment, and from industry, Carnegie Mellon University; City University of New York; Columbia University; University of Maryland; University of Massachusetts; Pennsylvania State University; Rensselaer Polytechnic Institute; University of California, Los Angeles; University of Aberdeen; University of Cambridge; University of Cranfield; Imperial College, London; Royal Holloway and Bedford New College (University of London); University of Southampton; and University of York.