Dissertation/Thesis Abstract

Enhancing Physical Layer Security in Wireless Communications Using Secrecy Extraction and Friendly Jamming
by Li, Hao, Ph.D., The George Washington University, 2020, 73; 28257892
Abstract (Summary)

Physical layer security has received much attention in the community. The physical characteristics of wireless channels, in particular the channel impulse response (CIR), have become an attractive source of shared secrecy between two communicating parties for many wireless applications. However, ensuring high-entropy secret bits from the CIR may not be always feasible, including in IoT and/or vehicle communications, where the coherence time may not be easy to obtain. Sometimes the mobility is limited. Many techniques have been developed to resolve the above problems, including privacy amplification for higher entropy output, and pre-processing for data distillation. Due to the nature characteristics of wireless communication, eavesdropping is always an issue in the area of security. Many studies have been focused on preventing eavesdropping attacks.

In this research, the above concerns are addressed. Meanwhile, a framework to extract entropy from sequentially arriving CIR observation samples is developed. Then, both online and offline algorithm are proposed. The optimal solutions are fully characterized through a series of analytic results, and the optimality of the proposed algorithms are established. The performance of the algorithms are verified through comparisons with the existing related techniques, and evaluated. Moreover, to deal with the eavesdroppers, a flex intermittent jamming approach is proposed. Which is optimal and make further improvement in reducing energy cost while preventing eavesdropper sniffing acute information. Then, an improvement is made by introducing support vector regression to predict the length of the packet that are going to be sent while alleviating the delay created by jammer accessing the length of the packet being sent.

The contributions of this dissertation research lie in the areas of physical layer security. This work has a formal proof of the entropy behavior. Our proposed algorithm holds significant promise to be applied in many applications. This research also provides an energy saving friendly jamming approach against eavesdroppers.

Indexing (document details)
Advisor: Choi, Hyeong-Ah
Commitee: Youssef, Abdou, Cheng, Xiuzhen, Arora, Arminder, Cheng, Wei
School: The George Washington University
Department: Computer Science
School Location: United States -- District of Columbia
Source: DAI-A 82/6(E), Dissertation Abstracts International
Subjects: Computer science, Information Technology, Technical Communication, Management, Artificial intelligence, Web Studies, Remote sensing
Keywords: Entropy, Friendly jamming, Physical layer security, Wireless channels, Channel implusive response, Wireless applications, Computer security , Internet of Things, Vehicle communications
Publication Number: 28257892
ISBN: 9798557003247
Copyright © 2021 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy