Dissertation Defence: Communication and Sensing in Ultraviolet Non-Line-of-Sight Omnidirectional Secure Transmission
May 1 at 12:30 pm - 4:30 pm
Tao Shan, supervised by Dr. Julian Cheng, will defend their dissertation titled “Communication and Sensing in Ultraviolet Non-Line-of-Sight Omnidirectional Secure Transmission” in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Electrical Engineering.
An abstract for Tao Shan’s dissertation is included below.
Examinations are open to all members of the campus community as well as the general public. Please email julian.cheng@ubc.ca to receive the Zoom link for this exam.
Abstract
Ultraviolet (UV) non-line-of-sight (NLOS) communication has emerged as a promising candidate for locally secure, omnidirectional communication due to its minimal ambient noise, robust weather and terrain adaptability, and inherent local confidentiality. However, most existing research on UV NLOS transmission remains confined to point-to-point systems, lacking efficient methods for rapid coverage estimation, standoff boundary quantification, and source localization in point-to-area scenarios. This thesis aims to investigate a comprehensive framework that supports both sensing and communication within UV NLOS omnidirectional links, thereby addressing the key challenges of coverage prediction, security analysis, and positioning under scattering-based transmission conditions.
We begin by developing a Monte Carlo integration model for coverage estimation in omnidirectional UV NLOS links, demonstrating a significant reduction in simulation time compared to classical methods. This model is then extended to quantify the system’s covert communication performance using newly introduced photon-count–based and peak-power–based metrics, enabling the derivation of standoff boundaries for secure transmissions.
Next, we explore how weather conditions—including aerosols, fog, and precipitation—impact NLOS coverage. Through a combination of Mie theory and classical atmospheric models, we characterize the scattering and absorption coefficients in various meteorological scenarios, revealing that certain conditions can unexpectedly enhance UV coverage ranges.
We subsequently focus on sensing in omnidirectional transmission, namely NLOS source localization. To facilitate NLOS source localization, we propose the use of a spatially resolved photon-level (SRPL) receiver and assess its performance through validation by an industrial Solar-Blind Photon-Level (SBPL) camera. Outdoor experiments confirm that when integrated with kernel density estimation based NLOS direction finding and received signal strength indicator based NLOS ranging, the SBPL camera localizes UV NLOS sources with a root-mean-square deviation of only a few meters in a \SI{6900}{\square\meter} test field.
Finally, we address inter-symbol interference (ISI) mitigation in NLOS UV communications using an SRPL receiver architecture. Our discrete-time finite-state model reveals that ISI photon-counting channels behave like multipath channels, where the path of greatest signal strength is not necessarily the shortest. We demonstrate that time-aligned equal gain selective combining, in combination with a counting instant that maximizes photon counts within a single window, concentrates the majority of signal photons into one counting interval and significantly enhances link reliability.
