Research by a South Dakota State University scientist and his colleagues elsewhere could make it safer to transfer information over mobile devices, such as cellphones, while conserving battery power, the researchers claim.

Researcher Wei Wang says those would be among the benefits from using a proposed new selective encryption technique to protect the most important content in streaming video over wireless networks. The technique helps guarantee security while using less energy.

Because the proposed plan would selectively encrypt information instead of encrypting everything, it would improve efficiency and reduce computing overhead.

Wei Wang, an assistant professor with SDSU's Department of Electrical Engineering and Computer Science, says those are the chief advantages of the technique.

Wang says the research addresses the needs for quality, security, and energy efficiency as the demand grows for video and other multimedia content that can be streamed over portable computing devices, which are getting smaller and smaller—easy enough to fit in a palm, in many cases.

"This is a trend: Computers get smaller and smaller, so the cost is reduced. However, because it is so small, it is not as powerful as a laptop or desktop. It has a limited battery and limited computing power. We want to achieve good quality and strong security with a limited battery and limited computing power. How do we do that? This is the focus of my research."

Wang and his colleagues started by examining the multimedia stream and determining that different parts of the data have different importance. That means that different parts of the data stream could be treated differently when it comes to allocating battery power and computing power.

"We determine the most important parts in the multimedia streams, and we try to serve the important data first with more resources to make sure the important data is correct and is transmitted with high quality," Wang says. "And for the unimportant part, we serve them with low quality. This method of unequal treatment is a new trend in computer networks and multimedia applications."

The research builds on past work by Wang and colleagues involving image encryption, but that work was directed at single images. Wang's earlier studies focused on the importance of encrypting position information as compared to value information—where something is in a photograph as compared to how dark or light it is, and where the darkness or brightness are, for example.

The earlier research provided solutions for helping to encrypt information when someone wants to use a mobile device for such tasks as scanning a personal check or making a bank deposit.

"The check may have some sensitive information, such as an account number. We don't have to hide all the information. Rather, we find out what is the most important or sensitive part of the data and then we encrypt those important data or blur those important data," Wang says. "So this is the concept behind our research: Find the locations, encrypt the locations, and the whole picture is hidden. By doing this, a significant amount of overhead in terms of encryption is reduced. The proposed encryption is faster and it's a natural fit for a low-cost, mobile computing device."