| Summary: | This thesis develops a framework for evaluating deep space communication architectures under uncertainty and explores the potential of optical communications to meet the future data requirements of deep space science missions. Currently, the Deep Space Network (DSN) relies solely on radiofrequency (RF) communications. As science-driven space exploration missions become more complex, the data requirements are pushing the limits of existing communications infrastructure. Incorporating optical communications into the network offers a way to achieve higher data rates by moving to a higher frequency band without increasing mass and power. However, optical communications require fundamentally different designs and infrastructure, necessitating substantial design and investment.
Aligned with NASA’s goals, the study sets a 100x improvement over current data rates as the performance benchmark for proposed designs. Five architectures, including arrayed ground stations, a relay at the Earth-Sun L4 Lagrange point, and a space-based receiver array in Earth orbit, are evaluated using data rate, availability, data volume, and cost proxy as figures of merit. A statistical model for data rate is developed, incorporating 22 input parameters to estimate laser data transmission using Pulse Position Modulation protocol.
The study develops a cost proxy model of component costs, revealing the cost-effectiveness of segmented mirror antennas. Evaluating 3000 designs for each planet scenario and architecture, statistical analyses determine the cost-optimal design achieving 100x data rate improvement at the lowest cost.
A Polynomial Chaos Expansion (PCE) is used to calculate the mean and variance of the data rate over a target planet’s orbit in approximately 1/2 the time of a Monte Carlo simulation. Results show that optimal designs vary across planet scenarios, with Venus-optimized designs featuring smaller fields of view and narrower bandpass filters due to higher background light conditions.
Each architecture offers distinct advantages and disadvantages in terms of cost, data rate, data volume, and availability, with arrayed architectures standing out as promising options for high data rates and reduced risk. The space-based receiver array enables average data rates of 256 Mbps from 1.74 AU (corresponding to an average Mars distancec), compared to 84 Mbps for a single ground station, with data volumes 8-10 times greater than a single ground station due to high availability and the absence of atmospheric effects. However, the single ground station remains the most economical choice in every scenario. Comparisons show that the baseline optical architecture for a single ground station is more cost-effective than large arrays of RF receivers for achieving similar performance improvements on Mars and Saturn. The nominal design for Mars requires 249 arrayed 12 m antennas at a total cost of $154.38 million, while the single optical ground station costs $97.18 million to achieve the same data rate.
The study also discusses the development and implementation of a novel hybrid RF/optical antenna at the Deep Space Network site in Goldstone, California. The system, featuring a 1.5 m segmented primary mirror, a superconducting nanowire single photon detector, signal acquisition and control cameras, and adaptive optics components, will receive downlink from the first demonstration of deep space optical communications from the DSOC payload on the Psyche spacecraft, scheduled to launch later this year.
|
|---|