Networks & Cybersecurity Research | ECE | Virginia Tech

Research Areas

In a world of cyberphysical systems networking research continues to evolve. Researchers work closely with wireless communications, but are also studying social networks and infrastructure networks such as the power grid. Our teams investigate issues of nodes and systems that relate to capacity, reliability, energy efficiency, resilience, and security.

Current Research

Monitoring algal blooms

Harmful Algal Blooms (HABs) in marine and freshwater environments are a global health and environmental concern. On-site sampling methods followed by in-lab analysis of HAB toxins are neither sustainable nor practical. Alternatively, identifying standard color products from satellite images is useful for monitoring general algal bloom activities, but cannot identify harmful blooms, nor indicate toxin release. We are investigating a complementary approach to the two different methods that uses an innovative wireless sensor network to monitor the level of toxins in situ and in real time.

Interference mitigation

Interference management and mitigation is a fundamental problem in networking. We developed an Interference Alignment (IA) algorithm for multi-hop underwater acoustic networks with large propagation delays. The algorithm jointly constructs the signals at transmitters, so that the constructed signals overlap at their unintended receivers but remain decodable at their intended receivers.

Critical infrastructure

Creating resilient smart cities requires the synergistic integration of cyberphysical critical infrastructures such as transportation, wireless communications, and water and energy networks. Such critical infrastructures have significant resource dependence as they share energy, computation, wireless spectrum, users and personnel, and economic investments. Their interdependence means they are prone to correlated failures due to day-to-day operations, natural disasters, or malicious attacks.

ECE teams are working to develop resilient processes that can control and manage these interdependent critical infrastructure resources. The research involves techniques from machine learning, operations research, behavioral economics, and cognitive psychology. We are also working to develop a new science of security that leverages synergies inherent in networked cyberphysical systems such as the smart grid and transportation systems.

Cognitive Security

Traditional authentication relies on cryptographic algorithms that verify user identity based on certain pre-configured secrets. In a world with trillions of mobile objects, rigorous key management techniques are difficult, the traceability of pre-configured secrets is compromised, and crypto-based authentication schemes become ineffective. ECE teams are investigating an alternative approach that supplements the crypto-based solutions with unique, unforgeable, and robust credentials that are inherent to the network.

For example, a location claim can be verified by a location "fingerprint" constructed from the ambient radio signals. A user's identity can be verified by knowledge from the user's online social networks. We are developing mechanisms to collect or learn these credentials through the normal network operations to securely verify a device or a user's identity or claims.

We have also developed new machine learning techniques that improve the security of cognitive radio networks against attackers trying to emulate primary, licensed users. Our work shows that, by performing device fingerprinting by using time-varying features, one can authenticate primary users and defend against primary user emulation attacks.

Maritime mobile

Marine communication technologies have limited capacity and are expensive to operate. We are developing long-range, self-powered ocean wireless communication links for marine broadband wireless communications. Compact, maintenance-free, low-cost wireless base stations can float in the water, harvest energy from waves and establish communication links with each other and to the Internet. This project will enable new maritime networked applications and impact ocean-related activities, such as fishing, recreational boating, marine transportation, oil and gas industry, ocean scientific study, and national security and defense.

Small cells

Cellular networks are shifting from a coverage-centric homogeneous deployment of high-power cell towers (base stations) to a more organic capacity-driven deployment that includes various types of low-power base stations, called small cells.

ECE teams are investigating a number of issues related to small cells, including distributed caching and optimizing resource management using machine learning, social ties and geolocation. We are also studying energy efficiency, resource usage, and large-scale dynamics while maintaining high quality in dense small cell networks. Another team is developing new analytical tools and metrics to establish performance limits for self-powered heterogeneous networks that use self-contained energy harvesting modules.

Virtualized networks

We are investigating fit-for-purpose wireless networks. We achieve this by focusing on characterizing the wireless network demand generated by different types of services and then seeking to dynamically construct networks to support these services. By modeling this process, we have shown that we can build networks that better satisfy users at lower cost than a more conventional approach.

HetNets Model

For analysis and academic research, very simplistic wireless network models are typically employed in order to maintain tractability, while for industrial design and development, complex system-level simulations with a very large number of parameters are generally used. This has made it difficult to estimate the actual gain that new techniques developed by researchers might provide in real systems. Our work has bridged this gap by developing new models with foundations in point process theory and stochastic geometry. These models are not only tractable but capture several non-idealities that are often ignored in other models.