Summary and Conclusion

In this work, we observe that future EO satellites will generate so much data that this data cannot be transmitted to Earth due to limited capacity of communication that exists between space and Earth. We showed that conventional data reduction techniques such as compression [130] and early discard [54] do not solve this problem, nor does a direct enhancement of today’s RF-based infrastructure [136, 153] for space-Earth communication. We explored an unorthodox solution instead - moving to space the computation that would have happened on the ground. This alleviates the need for data transfer to Earth. We analyzed ten non-longitudinal RGB and hyperspectral image processing Earth observation applications for their computation and power requirements and discovered that these requirements could not be met by the small satellites that dominate today’s EO missions. We made a case for space microdatacenters (SµDCs) - computational satellites tasked to support in-space computation of EO data. We showed that one 4KW space microdatacenter can support the computation need of a majority of applications. To address the communication bottleneck between EO satellites and SµDCs, we proposed three space microdatacentercommunication co-design strategies – 𝑘 −𝑙𝑖𝑠𝑡-based network topology, microdatacenter splitting, and moving space microdatacenters to geostationary orbit. These techniques enable effective usage of SµDCs.

在本文中，我们观察到，未来的地球观测（EO）卫星将产生海量数据，由于天地（space-Earth）通信容量有限，这些数据将无法完全传输回地球。我们证明了传统的数据削减技术，如压缩 [130] 和早期丢弃 [54]，无法解决这一问题，直接增强现有的基于射频（RF）的天地通信基础设施 [136, 153] 也同样无效。

为此，我们探索了一种非常规的解决方案 —— 将原本在地面进行的计算任务转移至太空。这减轻了将海量数据传输回地球的需求。我们分析了十种非时序性的RGB与高光谱图像处理地球观测应用，评估了它们的计算和功率需求，并发现当今主导EO任务的小型卫星无法满足这些需求。

因此，我们提出了建设太空微数据中心（SµDCs）的构想 —— 这是一种专用于支持EO数据在轨计算的计算卫星。

我们证明，一个4千瓦（KW）的太空微数据中心便可以支持大多数应用的计算需求。为解决EO卫星与SµDCs之间的通信瓶颈，我们提出了三种太空微数据中心与通信的协同设计策略

1. 基于 \(k-list\) 的网络拓扑
2. 微数据中心拆分
3. 将太空微数据中心移至地球静止轨道

这些技术使得对SµDCs的高效利用成为可能