Smart cities with integrated power grids, wireless communications, and intelligent transportation networks could improve everyday life for millions of people, but they would also be vulnerable to any number of cascading failures, due to natural disasters, day-to-day operations, or malicious attacks. These problems could lead to power outages and disruption of communications, which could bring an entire city to a screeching halt.

A multidisciplinary research team led by Walid Saad, assistant professor of electrical and computer engineering at Virginia Tech, has been awarded a $1.1 million grant from the National Science Foundation (NSF) to develop a framework that ties together techniques from network science, operations research and economics, machine learning, wireless communications, power systems, and psychology to develop processes that can give smart city systems the resiliency to recover from such failures.

The grant is part of a larger $2.5 million grant that is led by Virginia Tech and includes collaborators from Rutgers University and Florida International University. The Virginia Tech grant includes researchers from electrical and computer engineering, economics, and the Virginia Tech Transportation Institute.

"Laying the technological foundations of tomorrow's smart cities is one of the most critical challenges of the coming decade," Virginia Tech's research team wrote in their proposal.

Smart cities will require sharing and processing a huge amount of resources that are scattered across a number of critical infrastructures. These critical cyber-physical systems must work interdependently, opening many new points for failures that could easily spread to other areas.

The failure of a generator, for instance, could knock out power to residential areas, while also taking down wireless access points, causing a communications failure. Consequently, home customers, as well as connected vehicles that rely on the wireless network for control data, would be affected.

To prevent such failures from becoming catastrophes, the systems must be able to work together to manage common resources — energy, users, computational power, communication spectrum, etc. — to quickly recover.

The project goal is to develop a foundational framework to understand the interdependencies among a smart city's critical infrastructures and to design new, resilient resource management techniques that are cognizant of both the technology and the human users of this technology.

The interdisciplinary research will weave together notions from the study of cyber-physical systems, security, game theory, behavioral economics, psychology, transportation systems, social science, and power systems.

The framework will lead to theoretical and practical advances, such as the development of rigorous mathematical techniques for delineating the interdependencies between different critical infrastructures through a novel mix of graph theory and machine learning; new resilient resource management mechanisms based on game theory; and new behavioral models and studies related to trust relationships between a smart city's residents and critical infrastructure systems.

Saad has expertise in game theory, cyber-physical systems, and security. Others on the team include Sheryl Ball, an associate professor of economics; Danfeng Yao, an associate professor of computer science and head of the Human-Centric Security Laboratory; Myra Blanco, a human factors engineer at Virginia Tech Transportation Institute (VTTI); and Tammy Trimble, senior research associate at VTTI.

This smart cities grant is part of a larger NSF initiative known as Critical Resilient Interdependent Infrastructure Systems and Processes (CRISP). CRISP funds three- and four-year projects, each with funding of up to $2.5 million.

Saad also received a $150,000 grant from NSF, announced as part of the 2015 Global City Teams Challenge (GCTC), an activity launched in 2014 by the National Institute of Standards and Technology (NIST) to showcase smart technologies with the potential to transform cities and communities around the world.

Saad's GCTC grant is aimed at developing a machine-learning framework to enable the "Internet of Things" to dynamically identify, classify and authenticate devices based on their cyber-physical environment by leveraging data analytics tools. Saad is collaborating with Sanjay Raman from Virginia Tech's National Capital Region on this grant.

The idea is to develop environment-based device credentials that can verify a device's identity and validate the physical environment it claims to monitor and the actions it claims to be performing over time.

The research will involve high school students and science, technology, engineering and mathematics  educators, and deal with a variety of research topics, including security, cyber-physical systems and data analytics.

Share this story