Deep learning models errors in space memory for CubeSat missions
Injection of Space Radiation Errors on Low-power NVM using Generative Deep Learning
Hardware ArchitectureMachine Learning
Summary
Memory chips on small satellites can get corrupted by radiation in space, causing errors that are hard to predict. The authors studied how errors happen in a special type of memory chip used on a real CubeSat in low Earth orbit. They found that these errors often follow certain patterns and built a deep learning model that can generate similar error patterns. This helps engineers test software and hardware against realistic memory faults without needing more space missions.
What this means in practice
- •For satellite hardware designers: Simulate realistic radiation-induced memory errors to improve fault tolerance in CubeSat and small satellite hardware designs.
- •For embedded system developers: Generate plausible memory fault patterns for testing embedded software resilience against space radiation effects.
Tested on one dataset.
Authors
Eleonora Giudici, Antonio Miele, Luca Mottola
Abstract
We develop an error model for a Non-Volatile Memory (NVM) device that represents the effects of physical faults experienced in a CubeSat mission in Low Earth Orbit (LEO). The CubeSat carries an off-the-shelf low-power Ferroelectric RAM (FRAM) chip and runs repeated read/write tests according to a predefined pattern. We compare every memory snapshot received from the satellite with a reference bitmap to identify bit errors. We visually analyze erroneous values distributions in the memory snapshots and we understand that they are frequently distributed according to regular spatial patterns. We define a generative deep learning technique to automatically learn these spatial patterns and to synthetically generate additional error patterns mimicking the observed ones. The model is capable at producing memory errors for general application-level experimentation. Our preliminary validation indicates that when comparing the model output with the data we collect, 0.92 of the generated bit flips correspond to bit flips in the memory snapshots and the model captures 0.91 of the observed bit flips. A comparison of the spatial distribution of bit flips in 10,000 generated error patterns also provides a Wasserstein distance of 0.073, indicating that the two distributions differ by roughly 7% of the maximum possible displacement of their probability masses.