Abstract
An investigation of very small satellite miniaturisation techniques is presented, focusing on enabling sub-kilogram technologies targeted at space sensor network applications. Distributed space mission concepts are emerging for scientific and remote sensing applications requiring massively distributed systems, analogous to proliferating terrestrial wireless sensor networks. This particular architecture will enable observation of real-time multi-point phenomena. Space economics and environmental concerns dictate a cost-effective mass-producible low-mass satellite for brief but essential missions in low Earth orbit. Existing and emerging very small satellite technologies have been investigated, assessed, and compared, where power generation and payload volume are the key performance metrics. Two novel design methodologies have been developed, simulated, and verified through functional and environmental testing of hardware. SpaceChip, inspired by the satellite-on-a-chip vision, is a monolithic heterogeneous system-on-a-chip integration approach. SpaceChip proves widely applicable to sensor networks in hostile environments, including space, which require simple sensors and sub-kilometre separations. Five SiGe BiCMOS prototype chips have been fabricated which show promising results for two previously undeveloped subsystems. A method has been investigated for on-chip series connection of solar cells yielding a 3.4% efficient system-on-a-chip power supply. Furthermore, an environmentally-tolerant microprocessor design technique was developed that verifies the synergy of radiation hardening by design and asynchronous logic. PCBSat is proposed as a satellite-on-a-PCB miniaturisation approach focused on deriving the smallest practical satellite within the context of space sensor networks and constrained to the use of commercial components, processes, and deployment systems. The concept has been validated by flight model development and test, measuring 10x10x2.5 cm and 300 grams, for $10,000 to orbit in quantity. PCBSat emerges as an optimal tradeoff between cost and performance. A case study investigation of ionospheric plasma depletions, known to cause problematic navigation and communication outages, provided a comparison vehicle of all technologies considered in this effort. A demonstration mission based on PCBSat has been selected by NASA for launch in 2010. This research has advanced the state-of-the-art by providing new demonstrated cost-effective miniaturisation approaches enabling sensor network architectures.