CCI Funds 22 Cybersecurity Research Collaboration Grants

Project Title: 5GEM: 5G MEC-Enhanced C-V2X for Intersection Management

Principal Investigator: Duminda Wijesekera, professor, computer science, George Mason University 

Co-Principal Investigators: Jeffrey Reed, CCI chief technology officer, Willis G. Worcester Professor of electrical and computer engineering, Virginia Tech; Vijay Shah, research assistant professor, electrical and computer engineering, Virginia Tech; Steve Kan, director, CRASH Center, George Mason University; Bo Yu, post-doctoral associate, computer science, George Mason University

• Abstract: The primary goal of this proposal is to design, prototype and perform in-field experimental run of a 5G multi-access Edge Computing (MEC) server for Intersection management and the effect of age of information has on that design. Most CV2X and intersection management address motorized vehicles, and pedestrians at zebra crossings. But they do not consider other emerging modes of urban transportation such as bicycles and motorized scooters for short distance rides. One of the distinguishing components of our proposal is to develop an MEC for intersection management that includes motorized vehicles, pedestrian, bicycles and motorized scooters. Our usage scenario will receive the individual state information from all actors approaching an intersection, including their intended destination across the intersection and cellular broadcasting the most usable information for these three types of actors. To be useful, we will provide algorithms that can filter only the contextually relevant information to directly support their described destinations that minimize the delays at the intersections. We will develop in-lab, in simulation and in-field experiments that would be carried throughout the process to validate our designs, protocols and algorithms and test their viability, cybersecurity and performance in the real-world. We will collaborate with the Arlington County Department of Transportation in selecting the intersections of concern, and carry out the in-field testing at the Rosslyn headquarters.

Project Title: Collective and Collaborative Defense for Virginia Regional Cyber Ecosystems

Principal Investigator: J.P. Auffret, associate director, Center for Assurance Research and Engineering (CARE), George Mason University.

Co-Principal Investigators: Kevin Heaslip, associate professor, civil engineering, Virginia Tech; Abdul Rahman, CCI AI Testbed Director; Jamil Jaffer, executive director, National Security Institute, George Mason University; Karl Darin, VP, operations, ConnectedDMV; George Thomas, VP, innovation and strategic initiatives, ConnectedDMV

• Abstract: The Commonwealth of Virginia has enhanced its cybersecurity posture over the last five years but scaling challenges remain given Virginia’s varied regional critical infrastructure, supply chains and cyber ecosystems. In addition, the work from home environment resulting from the global COVID‐19 pandemic and increased cyber threat surface area due to the adoption of IoT devices and the rollout of 5G infrastructure are furthering the scaling challenge.  The project brings together Northern Virginia and Southwest Virginia CCI nodes with the George Mason Center for Assurance Research & Engineering, Volgenau School of Engineering, National Security Institute, Antonin Scalia Law School, Virginia Tech Hume Center for National Security and Technology, the CCI AI Testbed, and Connected DMV. The project proposes adapting the fusion center concept to regional collective and collaborative defense with Regional Cyber Centers (RCCs)  by developing innovative strategic, business, legal, policy, and operational constructs that are tailorable for a region’s cyber ecosystem.  The project will tailor the RCC model to the Greater Washington Metropolitan Area (DMV) and Blacksburg/New River Valley Planning District and provide a foundation for standing up the RCCs in 2022. The CCI AI Test Bed will develop and demonstrate  capability to assist RCCs in scaling including the application of AI‐ML to network sensor feeds for anomaly detection and threat identification; leveraging the Test Bed for both ingest of relevant data and usage of models for prediction, correlation, and analysis of  threats,  their severity, their impact; and enriching ingested data with other  local  data  to establish common operating pictures.

Project Title: Enhancing 5G Wireless Network Security with Reconfigurable Intelligent Surfaces

Principal Investigator: Kai Zeng, associate professor, Department of Electrical and Computer Engineering, George Mason University

Co-Principal Investigators: Brian Mark, professor, electrical and computer engineering, GMU; Lingjia Liu, associate professor, electrical and computer engineering, Virginia Tech; Changqing Luo, assistant professor, computer science, Virginia Commonwealth University

• Abstract: The concept of reconfigurable intelligent surface (RIS) has recently gained much research attention for beyond 5G technology applications. RISs are man-made surfaces of electromagnetic material that are electronically controlled with integrated electronics and have unique wireless communication capabilities. Although key benefits of RISs in beyond 5G applications have been demonstrated for enhanced capacity, spectral efficiency, and higher rates, its benefits for enhancing 5G wireless network security have not been well understood or explored. This project aims to build theoretical foundations and develop novel techniques for RIS-enhanced 5G wireless security that fully harvest the degrees of freedom provided by RISs. It consists of four research thrusts: 1) to develop a reinforcement learning (RL)-agent assisted physical layer key generation scheme for RIS-enhanced 5G wireless communications; 2) to investigate the use of LPD/LPI waveforms in conjunction with RIS to improve the security of 5G communications and increase spectrum utilization as a cognitive underlay; 3) to design efficient RIS-transmitter/receiver channel state information (CSI) estimation methods to support secure communication; 4) to build an RIS mmWave 5G testbed to evaluate and validate the proposed ideas. The proposed research is well aligned with the strategic focus areas at the CCI NoVa Node, including Strategic Focus Area 1.A.ii: Cyber Research for the National Defense and Initiative 1.B.i: 5G-enabled Test Beds. Part of the research results will be further developed as training and course materials. The developed testbed will also be made available to the CCI partners for training, teaching, and research activities.

Project Title: RSA-DC: Building Robust and Self-Adaptive Defense Capability in Cyber Systems

Principal Investigator: Songqing Chen, professor, computer science, George Mason University

Co-principal Investigator: Qun Li, professor, computer science, College of William and Mary

• Abstract: Most of our services today, ranging from the entertainment to the military, rely on various cyberinfrastructures. As such, attacks targeting cyberinfrastructures, and the corresponding damage and loss caused, keep increasing. While continuous efforts are being made to defend against such attacks, the arms race between attackers and defenders is getting more and more intensive. Most defenses are still mostly static. The efficiency and effectiveness of such defense systems are often heavily relying on the operator and domain experts and their knowledge and experience. Therefore, such systems often cannot keep pace with the attackers. This project proposes to build a RSA-DC, a framework that builds robust and self-adaptive defense capability for existing cyber systems. RSA-DC is built on top of the latest network technology and machine learning and AI techniques. To achieve this goal, the researchers plan to first leverage the software defined networking (SDN) and programmable data planes to achieve low overhead network event collector. Second, to minimize the delay and errors introduced by manual operations of operators, they propose to build a bias learning based dual reinforcement learning model to automate the operation of the systems based on the collected events. Such a model can self-adapt to the monitored events by guiding the operations of the defense system automatically. The researchers plan to build a prototype on cloud data center networks. Note RSA-DC does not exclude experts’ experience and domain knowledge. Instead, it aims to utilize such important knowledge more effectively in an automatic and self-adaptive manner.

Project Title: Spatiotemporal G-code Modeling for Additive Manufacturing Security

Principal Investigator: Rob Prins, engineering professor, James Madison University

Co-Principal Investigator: Irfan Ahmed, assistant professor, computer science, Virginia Commonwealth University

• Abstract: Smart manufacturing includes interpretation of digitally communicated designs into code intended for execution on connected devices such as those associated with additive manufacturing. Simulated cyberattacks on 3-D printers have demonstrated that mechanical properties of a part can be maliciously modified without causing defects that are apparent to the typical inspection process. Such concealed defects are of concern because they may result in parts failing after they are integrated into a larger system and deployed. Failure in this mode jeopardizes the entire system. Researchers have investigated ways to detect cyberattacks on manufacturing code (G-code) including methods that predict expected 3-D printer behavior based on known good code and compare these expectations to observed reality during the build process. The proposed research includes two thrusts. The first thrust will focus on development of several cyberattack simulations applicable to 3-D printers (including generation of internal voids and manipulation of process temperatures). The effect of these different attacks on mechanical properties will be observed through testing. The second thrust will focus on G-code modeling to predict printer behavior; G-code models will be compared to actual printer behavior using an integrity-checking framework. Actual printer behavior will be monitored via a novel technique that is more direct than methods currently described in the literature.

Project Title: User-Centric Privacy Controls for Smart Home Devices

Principal Investigator: Vivian Motti, assistant professor, information sciences and technology, George Mason University

Co-Principal Investigator: Carol Fung, associate professor, engineering, Virginia Commonwealth University

• Abstract: Smart home technologies include an ecosystem of devices aimed to provide convenient services for end users. Smart home devices (SHDs) vary on their form factor and purposes, serving as personal assistants for communication, entertainment, lighting, surveillance, among other services. SHDs are well-suited to automate services and manual activities, however due to their continued data collection, access to personal information from end users, and communication with third-party services through the web, these devices also threaten users’ privacy. Privacy concerns for smart homes are well-acknowledged in the scientific literature and popular media, however addressing privacy risks for SHDs remains an open challenge. To advance the research and development of user-centric controls for smart home devices, this project employs user-centric design. The project aims to systematically characterize privacy concerns from end users’ perspectives, formalizing privacy risks in a threat model for smart home devices. Also, the research team will define, develop and evaluate a privacy-enhancing framework for smart homes. The project contributions are three-fold: (i) a threat model of privacy risks for smart home devices; (ii) graphic user interfaces for privacy controls; and (iii) a catalog of privacy-preserving policies and design patterns. The project is in line with CCI goals investigating the impact of human behavior on cybersecurity focusing on privacy for smart home devices.

Project Title: A Systematic Evaluation of Smart City Security and Privacy

Principal Investigator: Adwait Nadkarni, assistant professor of computer science and director of the Secure Platforms Lab (SPL), The College of William & Mary.

Co-Principal Investigator: Yuan Tian, assistant professor, computer science, University of Virginia

• Abstract: Fueled by lessons learned from small-scale deployments (e.g., smart homes), we are beginning to see efforts at realizing the smart city vision—futuristic cities with various levels of autonomy and cyber-physical control over traditionally manual processes. However, the security and privacy implications of such large-scale deployments are understudied, with the primary challenge being their scale and complexity, caused by the deployment of devices and services with varied technology stacks, and for heterogeneous use cases, which is far beyond relatively small smart home deployments. This proposal seeks to bridge this gap by developing systematic methods to investigate the security and privacy of smart public spaces, i.e., larger deployments of IoT devices (e.g., in office buildings/streets) that constitute smart cities. We propose an empirical approach for evaluating smart public space deployments by formulating two research goals: (1) evaluating the security of network communication, and (2) analyzing the privacy implications of centralized management, particularly in terms of what administrators can observe, and how users perceive such observations. Our scalable research methodology systematically leverages well-founded techniques such as static and dynamic analysis, formal modeling, statistical language modeling, natural language processing, and user-studies. While our approach is general, we aim to evaluate it using the UVA Living Link Lab, a large-scale smart public space testbed. If successful, this proposal this proposal will develop a methodological understanding of the security and privacy challenges in smart city adoption, and an empirical foundation for future research on secure IoT communication as well as usable and privacy-preserving interfaces.

Project Title: Backdoor Detection and Mitigation in Deep Neural Networks

Principal Investigator: Hongyi Wu, Batten Chair of Cybersecurity and the director of the Center for Cybersecurity Education and Research, Old Dominion University

Co-Principal Investigators: Jin-Hee Cho, associate professor, computer science, Virginia Tech; Leigh Armistead, president, Peregrine Technical Solutions; Dave Wolfe, vice president, Peregrine Technical Solutions; Tom Murphy, Peregrine Technical Solutions; Chunsheng Xin, professor, electrical and computer engineering, ODU; Jiang Li, professor, electrical and computer engineering, ODU; Rui Ning, research assistant professor, cybersecurity, ODU

• Abstract: Artificial Intelligence (AI) is becoming a critical part of modern warfare, with its game-changing capability for handling large volumes of data and making collaborative complex decisions in support of self-control, self-regulation, and self-actuation of combat systems. While it is being adopted as an effective tool for automation where, relative to human intelligence, AI is faster, more agile, or low-cost, it is becoming an increasingly attractive target, opening vulnerabilities and enabling new forms of cyberattacks. In particular, a wide range of deep learning algorithms are vulnerable to polluted data, adversarial inputs, mimicry attacks, evasion attacks, and poisoning attacks. The resulting deep learning models are embedded with neural backdoors, that can cause catastrophic failure. The overarching goals of this proposed project include: (1) investigating existing and emerging neural backdoor attacks, (2) exploring and implementing effective backdoor detection schemes, and (3) once detected, developing and testing efficient backdoor mitigation techniques. Researchers from two CCI nodes (Coastal and Southwest) will work with defense industry and Navy/Army collaborators to develop the proposed backdoor detection and mitigation system, contributing significantly to the protection of future intelligent combat systems. The project will also develop new training modules in the context of secure and safe AI and offer them via workshops and bootcamps, aiming to prepare students and practitioners with advanced secure AI skills to succeed in cutting-edge research and industrial projects. Overall, the proposed work will lead to enabling technologies to secure AI systems, accelerating their development and widening their adoption in various application domains.

Project Title: DeepPOSE: Securing Transportation systems from GPS Spoofing Attacks

Principal Investigator: Chunsheng Xin, professor in the Center for Cybersecurity Education and Research, Old Dominion University

Co-Principal Investigators: Yanxiao Zhao, associate professor, electrical and computer engineering, Virginia Commonwealth University; Kevin Kefauver, technical director, Global Center for Automotive Performance Simulation; Danella Zhao, associate professor, computer science, ODU

• Abstract: Today the Global Positioning System (GPS) service is widely used in our daily lives. Smartphones, wearable devices, aircrafts, unmanned aerial vehicles (UAVs), self-driving cars or autonomous vehicles all rely on GPS to benefit from location based services, e.g., navigation, truck/vehicle monitoring, reporting self location in emergency or for rescue, searching nearby gas stations, restaurants, hotels, etc. While GPS becomes an indispensable element in our daily lives, it is rather vulnerable to GPS spoofing attacks even by low-cost hardware, resulting in potentially life-threatening impact. While there has been a long history of studying GPS spoofing, previous solutions to detect GPS spoofing were either applicable to limited scenarios only, or not effective to smart attackers. Furthermore, new challenges, such as varying attack surfaces and lack of mitigation techniques, make the problem even more challenging today. Capitalizing on a strong academia-industry partnership, a team of faculty from ODU and VCU and engineers from GCAPS VTT LLC proposes a holistic framework DeepPOSE to utilize multimodal sensor data to detect and mitigate GPS spoofing attacks to transportation systems, using Deep Learning technologies and emerging wireless communication techniques. The developed system will be prototyped and integrated into the product of GCAPS, a leading vehicle simulation framework, and results in potentially large economic impact on the Commonwealth of Virginia.

Project Title: Secure and Privacy Preserving 5G Network for Connected Vehicles

Principal Investigator: Sachin Shetty, associate director, Virginia Modeling, Analysis, and Simulation Center, Old Dominion University

Co-Principal Investigators: Cong Shen, assistant professor, electrical and computer engineering, University of Virginia; Duminda Wijesekera, professor, computer science, George Mason University; Kai Zeng, associate professor, electrical and computer engineering, GMU; Tan Le, research assistant professor, artificial intelligence, ODU; Jeffrey Reed, CCI's chief technology officer and the Willis G. Worcester professor of electrical and computer engineering, Virginia Tech

• Abstract: The integration of 5G and IoT technologies will result in connected vehicular networks that can profoundly change logistics management in the maritime and transportation industry. However, deployment of 5G empowered vehicular networks will also introduce attack surfaces targeting 5G-IoT ecosystem. This proposal aims to develop a secure and privacy-preserving 5G network capability in connected vehicular networks that can provide trusted information to maritime and transportation domains while protecting networked systems from being abused by malicious actors. Although there are several research efforts in 5G security, several research challenges that need to address to realize an integrated secure and privacy-preserving 5G-enabled vehicular network. In this proposal, we address specific challenges in securing 5G-enabled vehicular networks due to dynamic mobility, unknown and changing deployment environment, open wireless medium and limitations of centralized security infrastructure. Our specific differentiators include: (a) To maintain the effectiveness in changing deployment environment, federated learning can be automatically and continuously updated to fit the needs of specific deployment with no additional cost; (b) We provide an integrated security and privacy enhancing capability that will ensure maximum participation from end users; (c) Our Blockchain assisted distributed trust management employs advanced artificial intelligence approaches for data- and entity-centric trust models to ensure security of messages and vehicle trust; (d) Federated learning is integrated in our delay-aware CAN bus intrusion detection to build a robust and resilient distributed detection; (e) Finally, our solutions are deployed in resource constrained devices, such as, software defined radios, that will reduce cost while maintaining security effectiveness. 

Project Title: Virginia State Investments in Port of Virginia: A Simulation-based Framework for Identifying, Assessing, and Mitigating Systemic Cybersecurity at the Operational Technology Layer

Principal Investigator: Rafael Diaz, research associate professor, supply chain digitalization, Old Dominion University

Co-Principal Investigator: Haiying Shen, associate professor, systems & information engineering, University of Virginia

• Abstract: The modernization of terminal operations and facilities and the increased use of intermodal transportation at the Port of Virginia is illustrated in their Capital Investment Plans and the 2065 Master Plan. Significant investments are focused on electrification and promoting the use of alternative intermodal transportation. This transformation entails a shift towards the extensive use of operational technology (OT) and digitalization. The evolution toward cyber-physical systems (CPSs) increases the complexity of operations as devices extensively require reliable communication, coordination, and control systems. The emergence of CPSs and extensive use of OT inexorably increases operational complexity and exposure to cyberattacks. Regrettably, many OT systems are not designed against unprotected intrusion, and collectively, may represent a substantial source of supply chain disruption. These industrial OT systems are characterized by the presence of hierarchies and hyper-vulnerabilities at different operational levels. Cybersecurity protection of these systems requires a systemic approach to evaluate their vulnerability, cyberthreats, and countermeasures. However, current evaluating technologies are applied mainly to IT systems and commonly lack systemic perspectives that consider overlapping risks and tiered hierarchies at the OT layer. Particularly, as an extensive volume of vehicle movements occur at different stages of arrival, transit, and departure of cargo, false information attack of vehicle communication is of preponderant attention. With a view to overcoming these limitations, we propose to develop a simulation-based framework to gain a systemic cybersecurity perspective of the diverse and interconnected OT systems that define the CPS ecosystem of the Port of Virginia with a focus on vehicle communications.