Performance analysis and resource optimization in free space optical communication networks

Abstract

In today’s world, wireless communication has become a necessity due to the growing reliance on data-intensive Internet usage. The growing need for high transmission rates in bandwidth-intensive applications,such as ultra-high-definition video streaming, video-on-demand, and online video, further intensifies the current radio frequency (RF) spectrum shortage. In contrast, optical wireless communication (OWC) leverages unlicensed infrared and visible spectrums to offer gigabit-per-second data rates and provides substantial bandwidth capacity. Free space optics (FSO) communication is a kind of OWC technology that utilizes an infrared spectrum to transmit modulated light through free space using a coherent optical source. Compared to traditional RF systems, FSO offers several advantages, including access to unlicensed spectrum, ultra-high data rates,reduced electromagnetic interference, and flexible deployment. However, its effectiveness is constrained by the requirement for a clear line-of-sight (LoS) between transmitter and receiver.To address this limitation, this thesis investigates the use of an optical intelligent reflecting surface (OIRS), a passive beamforming device that enhances signal strength and mitigates LoS blockages. With their low cost,deployment flexibility, and range extension capabilities, OIRSs are poised to become integral components of future wireless networks. Despite the high data rate potential of FSO communication in next-generation net-works, it remains vulnerable to physical-layer security threats, particularly jamming attacks that can lead to denial-of-service (DoS). This thesis explores robust designs to improve the resilience and performance of FSO-based networks under such challenges.First, this thesis examines the performance of multiple possible integration architectures of the FSO/fiber-based back-end with the Light-Fidelity (LiFi)/Wireless-Fidelity (WiFi)-based front-end to deliver seamless connectivity to last-mile users. Link aggregation (LA) is also investigated to improve throughput and reliability. These proposed architectures are bench marked against conventional architecture in terms of throughput, cost-per-bit,fairness, and reliability.Next, the thesis proposes an integrated terrestrial-air-underwater (TAU) optical communication network and compares its performance with a traditional RF-based TAU network. A novel algorithm, COgnition-based Divergence Angle Tracking (CODAT) is introduced to dynamically optimize the divergence angle of the FSO beam, thereby improving data rate and minimizing outage probability.To tackle LoS challenges in multi-user environments, the thesis further explores the integration of unmanned aerial vehicles (UAVs) with OIRS-assisted FSO systems to support quality-of-service (QoS) requirements.Specifically, a mirror element assignment problem is formulated and addressed using both a heuristic approach and a sequential-fixing linear programming (SFLP) method. These solutions demonstrate superior performance in sum rate, fairness, and user coverage compared to benchmark strategies.Finally, the thesis presents a jamming-aware mirror element assignment scheme to mitigate physical-layer jamming in UAV and OIRS-assisted FSO networks. The proposed approach enhances system robustnes by improving sum rate, reducing blocking probability, and satisfying QoS constraints in adversarial environments.Overall, these contributions demonstrate the feasibility and effectiveness of robust FSO-based communication systems in various next-generation wireless network scenarios, offering significant advances in throughput, security, and deployment flexibility.