ORCID

https://orcid.org/0000-0002-0650-1987

Date of Award

Summer 2025

Embargo Period

7-31-2025

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

College/School/Department

Department of Nanoscale Science and Engineering

Program

Electrical and Computer Engineering

First Advisor

Dola Saha

Subject Categories

Electrical and Computer Engineering

Abstract

Over the years, the demand for wireless communications has grown exponentially, resulting in a scarcity of available spectrum. This scarcity leads to the need for efficient spectrum utilization, a key factor in ensuring reliable and uninterrupted communications. One of the key challenges in achieving reliable communication is ensuring secure transmissions to protect the privacy and integrity of data for active users. Wireless communications are vulnerable to eavesdropping, interception, and unauthorized access, which can compromise the confidentiality and integrity of transmitted data. The coexistence of multiple wireless communication systems is another challenge to be addressed for passiver users. In many cases, wireless communication systems have to coexist with passive users, which use the same spectrum for different purposes. Coexistence is especially challenging when at least one spectrum user is passive, as is the case with passive and active systems such as radio astronomy and wireless communication systems.

This thesis explores the efficient spectrum utilization from two aspects: secure communication between active users and coexistence of active and passive users in the same space, time and frequency. \textit{First}, we propose neural network-based approaches to enable secure communication and maintain the confidentiality of messages between intended users. Specifically, we present three approaches: 1) We design complex-valued neural networks that implement wireless steganography to conceal the existence of secret messages. Our method can achieve a high embedding rate and is independent of the waveform. 2) We propose an autoencoder-based approach for key generation, which extracts secret keys from compressed channel estimates. This approach can map non-reciprocal channel estimates to highly correlated keys. 3) We investigate the detection of unknown signals using two neural network architectures through multi-domain feature integration. Our models effectively distinguish between open-set and closed-set signals across a wide SNR range, outperforming approaches relying on single-domain features. \textit{Second}, we design a system that utilizes reconfigurable intelligent surfaces~(RIS) to establish coexistence and cancel interference at the telescope.

This encompasses several subtopics essential to the development and implementation of RIS, including a feasibility analysis of RIS design to evaluate its practical potential and effectiveness. The direction of arrival~(DoA) estimation in low SNR regimes is explored to provide critical data on incident interference for cancellation processes. Additionally, the prototyping of RIS arrays is addressed, focusing on the design of efficient arrays and the evaluation of performance against multiple DoA scenarios. These subtopics collectively contribute to advancing RIS technology for applications in wireless communication and interference cancellation. Our research contributes to achieving efficient spectrum utilization, resulting in improving the quality of signals received by radio telescopes and ensuring reliable and secure communications for active users.

License

This work is licensed under the University at Albany Standard Author Agreement.

Download

Share

COinS