DoD Grant Aims to Secure IoBT Networks Against Cyber Attacks)

Imadeldin Mahgoub, Ph.D., Tecore Professor and Director of Tecore Networks Lab in the Department of Electrical Engineering & Computer Science, was awarded a three-year, $591,173 grant from the U.S. Department of Defense (DoD) to acquire a deep understanding of the security challenges faced by Internet of Battlefield Things (IoBT) networks and utilize artificial intelligence (AI) to develop smart and effective solutions.

IoBT is a network of interconnected things specific to a battlefield environment, such as wearable devices, sensors, weapons, military robots, autonomous vehicles, and other objects that are self-configurable, self-aware, mobile, and dynamic. Military and defense organizations are showing increased interest in IoBT due to their ability to improve the effectiveness and efficiency of military operations.

However, IoBT networks operate in challenging battlefield conditions. They have no infrastructure and are highly unpredictable. The information is time-sensitive and decisions have to be made quickly. They have a high demand on bandwidth and are susceptible to outages as well as jamming by adversaries. In light of these challenges, securing an IoBT network against any type of cyber attack is critical.

Traditional security methods have developed various responses to cyber attacks, such as blocking an attack, and using honeypots, devices camouflaged inside the network that are designed to deceive a cyber attacker and gather information. Game theory has also been used to understand the attacker and formulate a response to attacks. Given the nature of IoBT and the unique security challenges it presents, there is a need for new intelligent methods, models, and schemes to address them.

“The goal of this project is to investigate the application of deception and camouflage using intelligent sensor gateways and honeypots to give the defender an advantage over the attacker in a game theoretical framework,” Dr. Mahgoub explained. “In our research proposal, we present a novel scheme that involves the use of AI, specifically machine learning (transfer and reinforcement learning) techniques, a network topology model incorporating dual function sensor gateways and honeypots, as well as a response system based on principles of game theory, to deceive attackers and efficiently detect, mitigate, and respond to cyber attacks on an IoBT system.”

There are several questions that Dr. Mahgoub and his team are striving to answer as they conduct their research. What is an optimal machine learning model for attack detection in IoBT systems? Where should such a model be located? Should it be a separate system or part of the sensor gateway? What game theory models best suit the IoBT environment?

The findings from this project will have significant impacts. The results are relevant to several DoD agencies, including the Army Research Laboratory, Office of Naval Research, and Air Force Office of Scientific Research, with various applications in the areas of cybersecurity, information assurance, and communications and networking. This information will also be integrated into vehicular networking mobile computing, IoT, and emerging network technologies research-oriented courses, and will be disseminated in educational and technical conferences and journals.

The grant will enable a postdoctoral fellow, several PhD students, and undergraduates to participate directly in the research and work on deception-based security of IoBT networks. In addition, dozens of graduate students will have the opportunity to work on related projects in Dr. Mahgoub’s existing courses on vehicular networking and smart mobile computing, as well as two new courses that focus on IoT and emerging network technologies.