Towards Highly Reliable and Efficient Space and Ground Network Connectivity

Abstract

Network connectivity forms the backbone of modern technological systems, facilitating data transmission and coordination across diverse domains of city, industry, agriculture, etc. In the context of diverse network deployments, this dissertation investigates two pivotal networks from the perspective of space and ground deployment, represented by satellite networks and LoRa networks. However, achieving highly reliable and efficient network connectivity remains a significant challenge due to inherent network properties and intricate deployment constraints.

This dissertation contributes towards highly reliable and efficient space and ground network connectivity. (1) Space networking connectivity: Typically, Earth observation satellites follow a rule of revisiting cycle to periodically pass over the same area of the Earth at regular intervals, which is uniquely determined by their orbital properties. However, this dissertation has discovered a phenomenon of revisiting cycle delay induced by partial Revisiting Imagery Delivery (RID), where the revisiting cycles of certain regions extend beyond their intended ones due to the transmission bottleneck in downlink bandwidth rather than the orbital properties. Therefore, this dissertation proposes Stride, an in-orbit edge computing system to redesign satellite transmission paradigms via in-orbit revisiting properties for satellite networks. Specifically, Stride exploits the inherent temporal redundancy in satellite imagery to enable space networking connectivity efficiency, and leverages the shared reference revisiting imagery for data reconstruction to enable space networking connectivity reliability. (2) Ground networking connectivity: The widespread deployment of unattended LoRa networks poses a growing need to perform Firmware Updates Over-The-Air (FUOTA). However, the FUOTA specifications dedicated by LoRa Alliance suffer from several deficiencies with respect to energy efficiency, transmission reliability, multicast fairness, and security. Therefore, this dissertation proposes FLoRa+, an energy-efficient, reliable, beamforming-assisted, and secure FUOTA system for LoRa networks. Specifically, FLoRa+ achieves processing gains by unlocking the potential of incremental update to improve ground networking connectivity efficiency, and achieves coding gain and beamforming-assisted power gain to improve ground networking connectivity reliability.

In conclusion, this dissertation advances the reliability and efficiency of space and ground network connectivity by architecting specialized end-to-end systems through in-orbit intelligence and hardware-software co-design, which drives the evolution of pervasive network connectivity to support the increasingly interconnected world.

Date of Award	16 Jun 2025
Original language	English
Awarding Institution	City University of Hong Kong
Supervisor	Weitao XU (Supervisor) & Kai LIU (Co-supervisor)