Wavelet researchers want to ease decision-making process

  • Published
  • By William J. Sharp
  • Air Force Office of Scientific Research Public Affairs
An Air Force Research Laboratory team began a new study on wavelets to help information analysts better prioritize workload. Basically, a wavelet represents a snapshot of information at a given point in time.

The research team's estimated $200,000, one-year-long project is titled: "A New Paradigm in Image Location -- Registration Using Polar Wavelets." Funding was provided by the Air Force Office of Scientific Research here.

Capt. Jamie Morrison is a member of the wavelet research team. He is an embedded systems design engineer with AFRL at Wright-Patterson Air Force Base, Ohio. Wavelets, he said, can be an important time saver for defense planners.

"For example, if you were doing an analysis of a radio station signal, a wavelet could be an analysis of one second of the signal," Captain Morrison said. "The one second of information represents something an analyst wants to study in detail."

If an analyst isn't interested in an eight-hour broadcast but is interested in certain sections of the broadcast, wavelets can help locate the information quickly and provide more information on the topic in which the analyst is interested.

With wavelets, the analyst must have reference points when making information requests. The better phrased the information request, the better analysis that can be done on a particular topic of interest.

"If you're looking for a certain airplane on an airfield but your search parameters are for the airfield, it may be difficult to find the aircraft you're looking for," he said. "However, if the search parameters are too narrow, you could find yourself focusing on an aircraft panel instead of the entire aircraft."

In some respects, wavelets are like data compressors. With enough trafficking of regular compressed files, such as on a network, information systems can overload, which could affect how quickly analysts can respond to problems. Wavelets, however, help analysts effectively manage the size and format of imagery allowing ease of use. Managing data in this way helps avoid overburdening the network while filtering out undesirable data.

"We're talking about huge networks -- not tens or twenties but thousands. On these networks, streams of text, audio, video and other information files are constantly running," Captain Morrison said. "With enough traffic, these files could choke a wireless network.

"Our research assumes sensors provide data analysts with large amounts of sometimes redundant information from multiple sensor networks," the captain said. "Our goal, then, is to help the analyst receive the information in a compact form so that military decision makers can focus on what interests them and make decisions on the best available data.

"We see wavelets as a method to help information analysts focus on the right information at the right time," he said.

By supporting research programs like this, AFOSR continues to expand the horizon of scientific knowledge through its management of the Air Force's basic research program. Many of the technological breakthroughs enjoyed by millions today, such as lasers, Global Positioning Systems and the computer mouse trace their scientific roots to research first funded by AFOSR.