Experimental subdiffraction source discrimination enabled by spatial demultiplexing and single-photon detectors

TL;DR

This paper demonstrates a universal method using spatial mode demultiplexing and single-photon detectors to discriminate faint sources beyond the diffraction limit, outperforming direct imaging especially in photon-starved conditions.

Contribution

The authors experimentally validate a parameter-independent test for asymmetric source discrimination that surpasses diffraction-limited direct imaging, accounting for realistic crosstalk and imperfections.

Findings

SPADE retains advantage over direct imaging at small separations and low intensity ratios.

An accessible crosstalk threshold of about 0.1 keeps false negatives below direct imaging.

SPADE requires up to ten times fewer photons than direct imaging for the same error rate.

Abstract

We experimentally demonstrate a universal, parameter-independent test for asymmetric source discrimination. The test allows us to discriminate faint sources well beyond the diffraction limit by exploiting spatial mode demultiplexing (SPADE) and single-photon detectors. Our test yields a rate of false negatives well below what can be achieved by diffraction-limited direct imaging. Our tabletop experimental setup is inspired by the problem of exoplanet detection, where one aims at detecting the presence of a faint source in the proximity of a brighter one. We present a complete theory, modelling arbitrary modal crosstalk, and collect data across a range of values for the source separations and intensity ratios. We show that SPADE retains an advantage over direct imaging in the relevant regime of small separations and low intensity ratios. Remarkably, we identify an experimentally accessible crosstalk threshold $C_{\mathrm{th}}\simeq 0.1$ below which the exponential rate of false negatives stays well below that of direct imaging. For example, for crosstalk of $10^{-2}$, SPADE needs up to one order of magnitude fewer photons than direct imaging to achieve the same error rate. These results demonstrate that SPADE offers an effective methodology for subdiffraction asymmetric hypothesis testing, under realistic imperfections and crosstalk, paving the way to photon-starved imaging tasks.