TinyML-Driven Cybersecurity for Autonomous Spacecraft: Latency-Accuracy Analysis for SPARTA RF and Cyber Threat Detection

2026-06-04

AI Summary

This study addresses the challenge of balancing low latency and high accuracy in onboard detection of RF-network integrated threats for autonomous spacecraft. Building upon the SPARTA attack model, it evaluates the latency–accuracy trade-offs of several TinyML-compatible models—including logistic regression, random forest, support vector machines, and multilayer perceptrons—against adversarial RF spectrograms generated by BandErasure, FakeNR, and NoiseBurst. For the first time, physical information-theoretic concepts such as Vapnik–Chervonenkis (VC) dimension and Lipschitz continuity are incorporated into spaceborne TinyML security modeling, yielding a lightweight baseline capable of microsecond-level inference. Experimental results demonstrate that logistic regression achieves only 1% lower accuracy than random forest while enabling microsecond-scale latency, validating its feasibility as an efficient onboard security baseline and highlighting future directions in multi-timescale learning and enhanced feature encoding.

Abstract

Autonomous spacecraft require rapid, lightweight, and reliable onboard detection of cyber-RF threats. Using the SPARTA attack model, we analyze the latency-accuracy trade-offs of TinyML-compatible classical models -- Random Forest, Logistic Regression, SVM, and MLP -- for detecting uplink jamming, Fake-NR spoofing, payload manipulation, ground-segment compromise, and unauthorized command injection. We present a physics-informed theoretical analysis of each model's computational complexity, VC dimension, Lipschitz continuity, and latency scaling, supported by empirical measurements on adversarial RF spectrograms generated via BandErasure, FakeNR, and NoiseBurst corruption modes. Results show that Logistic Regression achieves microsecond-level inference with only a 1\% accuracy drop relative to Random Forest, making it an effective TinyML baseline for onboard autonomy. The study also identifies opportunities for advancing spacecraft cybersecurity through richer feature encoders and multi-timescale learning architectures, building on recent progress in edge intelligence and trustworthy AI.

Problem

• autonomous spacecraft
• cyber-RF threats
• TinyML
• SPARTA attack model
• onboard detection

Innovation

• TinyML
• SPARTA
• latency-accuracy trade-off
• RF threat detection
• onboard cybersecurity