QUASAR: A Quantum-Classical Neural Network for SAR...

QUASAR: A Quantum-Classical Neural Network for SAR... | AI Research

Key Takeaways

QUASAR is a hybrid quantum-classical neural network designed to authenticate X-band Synthetic Aperture Radar (SAR) satellites by identifying unique hardware-...
X-band SAR satellites (8-12 GHz) play a critical role in disaster response, environmental monitoring, and military intelligence.
Yet, they lack robust physical-layer authentication (PLA), a security layer orthogonal to cryptographic solutions.
Existing PLA systems, typically based on radio-frequency fingerprinting, are often limited to sub-6 GHz frequencies and rely on classical deep learning.
However, this approach underfits the IQ phase nonlinearities that distinguish satellite hardware.

Paper AbstractExpand

X-band SAR satellites (8-12 GHz) play a critical role in disaster response, environmental monitoring, and military intelligence. Yet, they lack robust physical-layer authentication (PLA), a security layer orthogonal to cryptographic solutions. Existing PLA systems, typically based on radio-frequency fingerprinting, are often limited to sub-6 GHz frequencies and rely on classical deep learning. However, this approach underfits the IQ phase nonlinearities that distinguish satellite hardware. In this paper, we present QUASAR, to the best of our knowledge the first quantum-classical hybrid architecture that fuses a CNN spectrogram encoder with a variational quantum circuit (VQC) to provide PLA to X-band SAR signals. Our solution enjoys two distinctive features: (i) it is markedly more data-efficient than classical machine learning, requiring only 10% of the training data to match the accuracy of classical baselines -- data collection being notoriously the most time-consuming phase of PLA; and, (ii) at an equal data budget, it improves classification accuracy over those baselines. In detail, we test our solution under three adversarial scenarios: replay, crafted-IQ injection, and space-borne spoofing. QUASAR rejects spoofed transmissions in 89.7%, 94.1%, and 81.3% of attempts, respectively, establishing the first quantum-enhanced physical-layer classifier for satellite constellations. The fully detailed framework and the supporting results, other than being interesting on their own, show a novel research avenue for physical-layer authentication.

QUASAR is a hybrid quantum-classical neural network designed to authenticate X-band Synthetic Aperture Radar (SAR) satellites by identifying unique hardware-induced signal impairments. By combining classical deep learning with quantum computing, the researchers aim to provide a robust security layer that verifies the physical origin of satellite transmissions, protecting against spoofing and unauthorized signal injection.

Addressing X-Band Authentication

SAR satellites are critical for disaster response and military intelligence, yet they often lack robust physical-layer authentication (PLA). Existing methods typically rely on classical deep learning and are limited to sub-6 GHz frequencies. Because X-band signals (8–12 GHz) involve complex hardware nonlinearities, classical models often struggle to distinguish between satellites sharing the same hardware generation. The authors, Vincenzo Sammartino, Nathanael Denis, and Roberto Di Pietro, propose QUASAR to overcome these limitations by using a quantum-classical architecture that better captures minute phase perturbations and IQ imbalances.

The Hybrid Architecture

QUASAR fuses a classical convolutional neural network (CNN) with a four-layer Variational Quantum Circuit (VQC). The system uses an "IQ-native encoding" method that maps the amplitude and phase of complex IQ samples directly onto the polar and azimuthal angles of qubits. This approach preserves the information content of the signal, which the authors note is often lost in traditional real-valued encodings. By operating in a high-dimensional Hilbert space, the VQC can identify non-linear patterns in the signal that are difficult for classical models to detect.

Performance and Adversarial Testing

The researchers validated QUASAR using 3.76 TB of raw IQ data collected from 37 operational ICEYE satellites over 28 days. The study reports that QUASAR achieves 97.3% validation accuracy while using only 10% of the training data required by classical baselines. When tested against three specific adversarial scenarios, the system demonstrated the following rejection rates:

Replay attacks: 89.7% * Crafted-IQ injection: 94.1% * Space-borne spoofing: 81.3%

Research Implications

The results suggest that quantum-enhanced machine learning can significantly improve data efficiency and classification accuracy for satellite security. The authors note that their explainability analysis, which used gradient saliency maps, confirmed that the quantum branch of the architecture successfully amplifies hardware-specific fingerprints. While the study establishes a novel avenue for physical-layer authentication, the authors acknowledge that the framework requires periodic model updates via transfer learning to account for signal drift caused by the harsh space environment.