April 16, 2024
clear sky Clear 62 °F

NSF awards assistant professor for research to improve industrial wireless networks

Computer science faculty member Mo Sha to infuse artificial intelligence, machine learning to improve efficiency

Assistant Professor Mo Sha from Watson College's Department of Computer Science Assistant Professor Mo Sha from Watson College's Department of Computer Science
Assistant Professor Mo Sha from Watson College's Department of Computer Science Image Credit: Jonathan Cohen.

In the 21st century, manufacturing and processing plants increasingly rely on automation to reduce costs and keep things running smoothly. A sensor takes a reading, a controller decides the right course and an actuator carries out the task.

However, the two main industrial wireless standards that specify how to connect those sensors, controllers and actuators — WirelessHART and ISA100 — were developed nearly 15 years ago. Technological advancements have been rapid since then, and much of the day-to-day network management has been guesswork by human operators.

Assistant Professor Mo Sha is studying more efficient ways to run industrial wireless networks, and he recently earned a five-year, $500,000 National Science Foundation CAREER Award to fund his research. The CAREER Award supports early-career faculty who have the potential to serve as academic role models in research and education.

“When those industrial wireless standards were developed, even the research community did not have full knowledge of how we should do things,” Sha said. “They relied on what they had in 2007 and came up with many solutions based on research telling them that it was probably the best way to do that.”

A faculty member in the Department of Computer Science at Binghamton University’s Thomas J. Watson College of Engineering and Applied Science, Sha earned his bachelor of engineering degree at Beihang University and his master of philosophy degree at City University of Hong Kong (both in his native China) before receiving his PhD from Washington University in St. Louis. He joined the Binghamton faculty in 2015.

He has made wireless networks and how they have evolved, particularly in industrial settings, one of his main research areas.

“Traditionally, 40 or 50 years ago, a manufacturer would use cables to connect everything, but cables are too expensive to deploy and maintain, and they are very inconvenient when you want to add a new process requirement,” he said. “So now companies are trying to use wireless technology to replace all those cables.

To fulfill his CAREER Award, Sha is hoping to utilize machine learning to provide new methods to optimize network performance.

“Most networks are configured largely based on experience,” he said. “In a lot of scenarios, the network is not configured optimally. This research is trying to replace those manual, experience-based methods with more rigorous and scientific methods. We propose to use machine learning and wireless technology to come up with new solutions to configure the networks automatically. We’ll rely on the theoretical models and simulation methods that have been developed by the research community in the last 20 years.”

By cutting down on human error, Sha sees networks that are more efficient, more secure and better at responding to problems. There also could be economic benefits for companies adopting such networks.

“I’m hoping that the outcome of my research can contribute to the post-pandemic recovery for the manufacturing and processing industries,” he said. “It’s been so hard for them because of social distancing, and some plants may be shut down due to high infection rates. This new tactic will decrease human involvement in the network management process, so it’s better for social distancing and it reduces operating costs.”

He added: “More importantly, after having more advanced wireless networks, we can develop new applications that can’t be done with the current networks. We can further enhance our efficiency and make the U.S. more competitive.”

Sha’s NSF CAREER Award is titled “Advancing Network Configuration and Runtime Adaptation Methods for Industrial Wireless Sensor-Actuator Networks” (#2046538).