Project web site: NSF-2211616, NSF-2211617, NSF-2211618

Overview

The use of frequencies above 100 GHz for wireless links is rapidly emerging as a key enabler for future (beyond 5G) wireless systems. These high-frequency communications systems, which are referred to as terahertz (THz) links, offer numerous exciting possibilities, such as plentiful bandwidth for ultra-high speed data transmission, high-resolution sensing, and enhanced resilience against malicious attacks such as eavesdropping and jamming. Yet, so far, little research has been devoted to the question of how to implement a network that can provide high bandwidth links for multiple mobile users, while still maintaining security against eavesdroppers. This joint design problem remains unaddressed.

The objective of this collaborative NSF-funded research project is to realize physically secure, networked (mobile and multi-user), spectrally efficient, broadband links at frequencies above 100 GHz. Our team is developing a set of methodologies to enable (1) rapid localization of users in a mobile network environment including both friends and foes, (2) establishment of multiple directional links which simultaneously optimize data throughput and security, and (3) discovery and mitigation of eavesdropper attacks, including both attacks on data links and those which attempt to leverage control-plane functions which leak channel information.

Team

Daniel Mittleman, Brown University

Edward Knightly, Rice University

Kaushik Sengupta, Princeton University

Intellectual Merit

The proposed research program is among the first comprehensive study of terahertz wireless networks which holistically incorporates considerations of network functionality and security. The overarching goal of this project is to develop a radically new node architecture which can intrinsically support multiple access for mobile clients in a broadband THz network, while also maintaining a high degree of security against malicious eavesdropping. One thrust of this project involves the exploration of novel antenna designs which exploit strong angular dispersion. We propose a new method to enable active fast electrical tuning of such devices. This approach, uniquely suited to the THz spectral range, will be exploited for detection of an eavesdropping attack, as well as for localization of legitimate users and mobility detection. A second project thrust targets to develop spatio-temporal modulated array architectures for scrambling the information contained in side-lobes of the broadcast, via spectral aliasing. Combined with agile sensing functionalities, the proposed interface will selectively create secure zones for communication. We will explore the limits of performance of these systems with respect to spectral efficiency, quantify the concept of information beam width and consider the implications for various forms of malicious attacks. In a third thrust, we will leverage the power of this new node architecture to ensure that the quality of service for multiple mobile users is maintained, even while guaranteeing that eavesdroppers are unable to access, not only the primary communication channels, but also control plane functions required to establish and maintain mobile links. We will consider a powerful adversary, with multiple eavesdroppers employing machine learning capabilities, in a multi-user mobile network. The result will provide the optimal performance for a spectrally efficient and secure THz network, even in the presence of a sophisticated attack by colluding eavesdroppers.

Broader Impacts

The proposed research will lay the foundations for future networks operating above 100 GHz. These results will be the first discussion of the issue of security as a key design parameter in the development of THz networks, and will therefore have a significant impact on policy makers at FCC and internationally. The results will also impact standards bodies such as IEEE and ITU, by emphasizing the need to consider security side by side with functionality in the design of all layers of the network.

Publications

Z. Shaikhanov, S. Badran, J. M. Jornet, D. M. Mittleman, and E. W. Knightly, “Remotely positioned
metasurface-drone attack,” in Proceedings of the 24th International Workshop on Mobile Computing
Systems and Applications (HotMobile), 2023, pp. 110–116. DOI: 10.1145/3572864.3580343

B. Bilgin and E. Knightly, “Metashield: A multi-function AP-surrounding metasurface,” in Proceedings
of the 17th ACM Workshop on Wireless Network Testbeds, Experimental Evaluation & Characteri-
zation (WINTECH), 2023, pp. 1–8. DOI: 10.1145/3615453.3616510