Photonic kernel machine learning for ultrafast spectral analysis

TL;DR

This paper presents a novel photonic kernel machine approach for ultrafast spectral analysis of noisy radio-frequency signals, combining machine learning versatility with photonic hardware speed for high-throughput applications.

Contribution

It introduces a new photonic kernel machine scheme for spectral analysis, demonstrating theoretical principles and numerical performance on various signal processing tasks.

Findings

Achieves high throughput in spectral analysis of RF signals.

Resilient to noise and requires minimal control in photonic implementations.

Effective on both pulsed and continuous RF signals.

Abstract

We introduce photonic kernel machines, a scheme for ultrafast spectral analysis of noisy radio-frequency signals from single-shot optical intensity measurements. The approach combines the versatility of machine learning and the speed of photonic hardware to reach unprecedented throughput rates. We theoretically describe some of the key underlying principles, and then numerically illustrate the reached performances on a photonic lattice-based implementation. We apply the technique both to picosecond pulsed radio-frequency signals, on energy-spectral-density estimation and a shape classification task, and to continuous signals, on a frequency tracking task. The presented optical computing scheme is resilient to noise while requiring minimal control on the photonic-lattice parameters, making it readily implementable in realistic state-of-the-art photonic platforms.