University of Nebraska–Lincoln faculty continue to make strides in bridging the digital divide between rural and urban America through innovative research, this time with a fellowship and new collaboration with another Midwest university.
With support from the National Science Foundation and its Established Program to Stimulate Competitive Research Fellows initiative, Qiang Liu, assistant professor in the School of Computing, will join researchers at Iowa State University to develop a first-of-its-kind, zero-touch network management system with a new, safe, online hierarchical learning framework for open radio access mobile networks.
The EPSCoR Research Fellows program is designed to support early-career research investigators on extended and collaborative visits to private, government, or academic research centers, such as another partnering university or national laboratory.
“They gain experience, work with the best technology, use their advanced infrastructures, do state-of-the-art research, and then bring learned knowledge back to their home institute,” Liu said. “It benefits the principal investigator as well as the university and the state.”
Building on his previous wireless network research, Liu’s new project will focus on the development of autonomous mobile 6G networks. He'll design novel artificial intelligence and machine learning techniques to address real-world network management challenges such as safety, scalability, robustness and practicality.
Liu said he chose to conduct his research at Iowa State since the university shares many research focuses with the University of Nebraska–Lincoln. While most other EPSCoR partner universities are located in urban areas, Nebraska and Iowa State’s Midwest locations and strong state agricultural industries make them prime places to conduct rural connectivity and application research. Iowa State’s unique city-scale network infrastructure offers additional opportunities for groundbreaking wireless network research.
While a base station providing wireless coverage will typically cover a radius of a few miles, Iowa State’s infrastructure includes several base stations located throughout its city of Ames. While multiple base stations can provide a broader coverage area, they can also create wireless interference issues. According to Liu, potential interference is actually advantageous for his research, allowing him to develop more robust, adaptive algorithms and mitigation solutions.
“Interference management is one of the big issues of wireless networks,” Liu said. “With this unique infrastructure, we can test interference and make sure that our AI machine learning model is learning, and we can apply that to a city-scale model rather than a lab-scale model.”
With the ability to test in a more realistic setting, Liu will be able to deploy and refine both real-time and non-real-time resource allocation and network configuration through online hierarchical learning techniques.
“AI machines learn to adapt, but when two of them are learning at the same time, they could conflict and make their system unstable,” Liu said. “We address that problem with online hierarchical learning, which optimizes the network by making sure they are well informed with each other in different approaches, not conflicting with each other, and autonomously learning by themselves without diverging from the system or making it unstable.”
Liu will begin his research at Iowa State this summer and continue to expand on it at Nebraska through his ongoing project, Husker-Net, the University of Nebraska–Lincoln’s own private 5G network. Husker-Net is the first private 5G network in the state of Nebraska, and one of the only private 5G university research networks in the country.
Liu believes that this EPSCoR fellowship research will not only improve the functionality of Husker-Net, but also serve as a blueprint for advancing and improving networks across the country. The use of artificial intelligence and machine learning techniques for mobile network management have the potential to radically increase efficiency and reduce operating expenses for next-generation networks. These benefits would facilitate more widespread adoption in networks across Nebraska, the Midwest, and other partnering EPSCoR locations.
Liu said the NSF EPSCoR fellowship and partnership opportunity will offer many benefits to Nebraska and Husker-Net, as well as many other researchers in the future.
“That's the effort of NSF. They invest and build something so that other universities and researchers can use it instead of building their own, and then because they have more resources there, they can build it at a scale, rather than having everyone cooking in their own labs,” Liu said. “We can learn from them to improve our research capabilities, bring the knowledge back, and then enhance the work we are doing as well as Husker-Net.”
