We report a measurement workflow free of systematic errors consisting of a reconfigurable photonnumber-resolving detector, custom electronic circuitry, and faithful data-processing algorithm. We achieve unprecedentedly accurate measurement of various photon-number distributions going beyond the number of detection channels with average fidelity 0.998, where the error is contributed primarily by the sources themselves. Mean numbers of photons cover values up to 20 and faithful autocorrelation measurements range from g (2) = 6 × 10 −3 to 2. We successfully detect chaotic, classical, non-classical, non-Gaussian, and negative-Wigner-function light. Our results open new paths for optical technologies by providing full access to the photon-number information without the necessity of detector tomography.

Light is an essential tool for connections between quantum devices and for diagnostic of processes in quantum technology. Both applications deal with advanced nonclassical states beyond Gaussian coherent and squeezed states. Current development requires a loss-tolerant diagnostic of such nonclassical aspects. We propose and experimentally verify a faithful hierarchy of genuine n-photon quantum non-Gaussian light. We conclusively witnessed 3-photon quantum non-Gaussian light in the experiment. Measured data demonstrates a direct applicability of the hierarchy for a large class of real states. PACS numbers: 42.50.Xa, 42.50.Ar, 42.50.Dv Individual photons as bosonic elementary particles have been subjects of a detailed quantum analysis already for many decades. It is intensified now due to their importance for quantum technology. First, a single photon antibunching was measured as incompatible with classical coherence theory [1,2]. It was the first proof of nonclassical light. This measurement became canonical for single photon sources [1][2][3][4][5]. After many years, broadband homodyne detection allowed indirect estimation of their continuous variable nonclassical features [6][7][8][9][10][11]. Their visualisation in the phase space of continuous amplitude of the electric field by a Wigner quasiprobability distribution shows multiple negative concentric annuli for Fock states of light [12]. The Wigner function is used to distinguish different Fock states of light, however, without any proof yet that they really form a faithful hierarchy. A faithful hierarchy of n-photon quantum non-Gaussianity would reliably recognize that, for a given order n, an observed state is statistically incompatible with any mixture of Fock-state superpositions up to |n − 1 modified by an arbitrary Gaussian phase-space transformation [10,13,14]. The hierarchy is schematically presented in Fig. 1. Unfortunately, such a faithful hierarchy based on the negative parts of the Wigner function has not been discovered yet and it would be anyway applicable only if overall losses were below fifty percent [15]. Since a large variety of experimental platforms emitting or transmitting light does not suppress the losses so much, a lack of theoretical tools witnessing genuine n-photon quantum non-Gaussianity limits optical diagnostic of quantum processes in matter, current fast development of multiphoton sources and their applications in quantum technology.A large gap between basic nonclassical light and light with a negative Wigner function was partially covered when a loss-tolerant direct measurement of single-photon * lachman@optics.upol.cz |4⟩ |3⟩ |2⟩ |1⟩ |0⟩ FIG. 1: A visual presentation of the hierarchy of genuine quantum non-Gaussian states approaching ideal Fock states of light. The white regions stand for mixtures of Gaussian states (squeezed coherent states). All colored regions represent states beyond those mixtures. Each color corresponds to a new quantum feature attached to highly nonclassical states such as Fock states |n (green points...

We propose an experimental method of recognizing quantum non-Gaussian multiphoton states. This is a native quantum property of Fock states, the fundamental quantum states with a constant number of particles. Our method allows experimental development and characterization of higher Fock states of light, reaching even beyond the current technical limits of their generation. We experimentally demonstrate that it is capable of distinguishing realistic quantum non-Gaussian light with the mean number of photons up to five despite detection efficiency of 50%. We also provide evidence that our method can help to distinguish the number of single-photon emitters based only on their collective emission.

Quantum oscillators prepared out of thermal equilibrium can be used to produce work and transmit information. By intensive cooling of a single oscillator, its thermal energy deterministically dissipates to a colder environment, and the oscillator substantially reduces its entropy. This out-of-equilibrium state allows us to obtain work and to carry information. Here, we propose and experimentally demonstrate an advanced approach, conditionally preparing more efficient out-of-equilibrium states only by a weak dissipation, an inefficient quantum measurement of the dissipated thermal energy, and subsequent triggering of that states. Although it conditionally subtracts the energy quanta from the oscillator, average energy grows, and second-order correlation function approaches unity as by coherent external driving. On the other hand, the Fano factor remains constant and the entropy of the subtracted state increases, which raise doubts about a possible application of this approach. To resolve it, we predict and experimentally verify that both available work and transmitted information can be conditionally higher in this case than by arbitrary cooling or adequate thermal heating up to the same average energy. It qualifies the conditional procedure as a useful source for experiments in quantum information and thermodynamics.

This work presents stochastic approaches to model the counting behavior of actively quenched single-photon avalanche diodes (SPADs) subjected to continuous-wave constant illumination. We present both analytical expressions and simulation algorithms predicting the distribution of the number of detections in a finite time window. We also present formulas for the mean detection rate. The approaches cover recovery time, afterpulsing, and twilight pulsing. We experimentally compare the theoretical predictions to measured data using commercially available silicon SPADs. Their total variation distances range from 10 −5 to 10 −2 .