Wednesday, June 25, 2014

CyLab Researchers In 5-Year NSF-Funded Project to Overcome Challenges in Synchronizing Time for Cyber-Physical Systems

A Sundial in Saint-Rémy-de-Provence, France (Source: Wikimedia)

NOTE: This CyLab Chronicles is cross-posted on both the Carnegie Mellon CyLab public site and the CyLab partners-only portal. - See more at:

The National Science Foundation (NSF) recently announced "a five-year, $4 million award to tackle the challenge of synchronizing time in cyber-physical systems (CPS)--systems that integrate sensing, computation, control and networking into physical objects and infrastructure." According to the NSF, the award "will support a project called Roseline, which seeks to develop new clocking technologies, synchronization protocols, operating system methods, as well as control and sensing algorithms ... Examples of cyber-physical systems include autonomous cars, aircraft autopilot systems, tele-robotics devices and energy-efficient buildings, among many others." (NSF, 6-13-14)

Two CMU CyLab researchers, Raj Rajkumar and Anthony Rowe, are members of the Roseline project team. Mani Srivastava of UCLA and Rajesh Gupta of UC San Diego will serve as Principle Investigators. The team also includes Sudhakar Pamarti of UCLA, João Hespanha of UC Santa Barbara and Thomas Schmid of the University of Utah.

CyLab News caught up with both CyLab researchers to get their perspectives on the scope and significance of Project Roseline.

"We all know that time is a fundamental attribute," Rajkumar said. "Computers maintain time by using local clocks, and synchronize these values among themselves or with reliable clock sources on the internet. However, the accuracies of these synchronized clocks are very highly system- and network-dependent. This in turn causes a wide range of applications from smart grid systems to robotic systems like autonomous vehicles to be customized and tweaked. In other words, there does not yet exist core foundations to specify, implement and utilize a notion of time, whose accuracy can be specified, controlled and achieved - we refer to this as the Quality of Time. This project will develop the foundations for managing the Quality of Time in computer systems." 

"There is a notion of time that transcends all layers of modern computer systems," Rowe adds. "At the lowest-level you have clocks driving hardware. Above that you have operating systems and networking that use time to manage how and more importantly when resources should be consumed. At the very top you have applications ranging from GPS to banking transactions that rely on timing. Our goal is to develop mechanisms and interfaces for improving what we call Quality of Time (QoT) aware applications. Specifically at CyLab we will be working on operating system abstractions and approaches to network coordination that improve energy-efficiency and reliability of networked embedded systems. Our target applications range from secure smart-grid monitoring to robotic systems like autonomous vehicles."

Full Text of NSF Press Release