smart plug research

Joey Taylor works on his summer undergraduate research project at WSU.

PULLMAN, Wash. – When Joey Taylor signed up to continue his research on smart plugs this summer, the plan was to spend most of his time programming and testing the device’s ability to transfer data over power lines.
As it often goes in research, however, he ended up spending about two months reverse engineering the plug he was given to test.
“Figuring out how the plug worked was … well … an event,” said Taylor, a junior in the Washington State University School of Electrical Engineering and Computer Science (EECS). A hearty laugh came from student Vignesh Ramachandran, who nodded in agreement as he sat next to Taylor in the computer lab where the pair spent most of their summer.
Plugs monitor power usage
They met about 10 weeks ago through the National Science Foundation (NSF)-funded research experience for undergraduates (REU) led by Nirmalya Roy, a clinical assistant professor in EECS. Ramachandran, like most of the other REU students, is from out of state; he is an undergraduate at San Jose State.
NSF funds REU programs all over the country as a way of introducing undergraduates to research and providing a unique, out-of-classroom and networking experience.
Taylor and Ramachandran’s project is part of Roy’s ongoing research on smart plugs, devices that monitor how much power appliances are using. Taylor began working with Roy on the research under a different NSF grant in the fall.
Transmitting data via power lines
“Smart plugs are useful to help businesses and other property owners track where the most power is used in their buildings and to find ways to save energy,” Taylor said.
The research aims to transmit data from a smart plug to a computer using power lines instead of wireless networks.  Taylor, who began teaching himself programming in middle school, was asked to program a smart plug and find out if it is possible to transmit data over power lines.
“Since the plug is already using power lines, transmitting the data that way would mean you don’t have to install wireless infrastructure or saturate a building’s broadband,” Roy said.
Unraveling software info first
However, Taylor did not get to that point in his research until about a week before the end of the REU program. The software on the smart plug Taylor was testing couldn’t be compiled. The plug came from another university that had stopped working on the project, and the existing software had been broken by upgrades to other parts of the operating system.
Additionally, the plug did not come with a reference sheet, so before Taylor could answer his question, he needed to unravel exactly how the computer chips in the plug were communicating, collecting data and working together.
Essentially, the inner workings of the smart plug were like pages torn out of a book and thrown to the wind. The pages of information were mixed up or missing all together – and in a foreign language. Taylor’s task was to interpret the pages and fit them together again so they communicated the correct message.
Intensive learning
After extensive work to understand the complicated smart plug communications, Taylor and Ramachandran were finally able to get to the problem they set out to solve a week and half before their project ended. The wrote new software during 12-14 hour days.
“I’ve never worked so intensely on one project before. I probably learned more in those two weeks than I’ve ever learned in any other two week span,” Ramachandran said.
Just two days before the final poster session for WSU summer research students, Ramachandran was interpreting the data and could identify four of the six devices that were plugged into the smart plug. They presented the research at WSU’s Undergraduate Research Poster Symposium.
“Joey is one of the most talented undergraduate students I have ever seen,” Roy said. “He was successful in collecting data over power lines, something I’m not sure Ph.D. students have been able to do yet.”