

Faculty Profiles
“A complicated problem might have simple solutions,” said Guoliang Xue, a professor in the Department of Computer Science and Engineering. “We just haven’t found them yet.” Xue is referring to his research in computer networks, specifically the work he is doing in multi-constrained Quality of Service(QoS) routing. He describes QoS routing as “finding a routing path that meets certain system requirements known as ‘quality of service.’” Such requirements are usually related to the cost or reliability of data packets.
Xue’s goal is to design simple algorithms that are “provably good” and applicable to wireless networks. Right now, most approaches to multi-constrained QoS routing tend to be either simple and speculative or sophisticated, time-consuming and expensive. The real difference between the two? Xue compares it to taking a road trip, in that there are maps and routes that a driver can locate in minutes that might work out just fine in most cases. Yet, they might also fail to caution the driver about roadwork or other obstacles and delay the trip. Similarly, uncomplicated algorithms are easier and less expensive to implement, but they might not provide any performance guarantee.
Good algorithms for multi-constrained QoS routing are extremely important in applications such as video streaming and Voice over Internet Protocol, the routing of voice conversations over the Internet. As a result, this problem has been extensively studied by researchers in the networking community. In a sequence of papers published in the IEEE/ACM Transactions on Networking, Xue and his collaborators have presented the fastest approximation schemes for these problems. “The best solutions will be those that are simple to understand, simple to implement and that are provably good,” said Xue, “If successful, the resulting algorithms can be implemented in network routers.”
Xue has published over 160 refereed papers and serves on the editorial boards of several premier networking journals.