Search

Gaussian states are ubiquitous in quantum optics and information processing, and it is essential to have effective tools for their characterization. One such tool is a photon-number-resolving detector, and the simplest configuration involves counting the total number of photons in the state to be characterized. This motivates the following question: What properties of a multi-mode Gaussian state are determined by the signal from one detector that measures total number photons up to some bound? We find that if the Gaussian state occupies $S$ modes and the probabilities of $n$ photons for all $n\leq 8S$ are known, then we can determine the spectrum of the Gaussian covariance matrix and the magnitude of the displacements in each eigenspace of the covariance matrix. Nothing more can be learned, even if all photon-number probabilities are known. When the state is pure, the covariance matrix spectrum determines the squeezing parameters of the state.

Randomized benchmarking (RB) is a widely used strategy to assess the quality of available quantum gates in a computational context. RB involves applying known random sequences of gates to an initial state and using the statistics of a final measurement step to determine an effective depolarizing error per step of the sequence, which is a metric of gate quality. Here we investigate the advantages of fully randomized benchmarking, where a new random sequence is drawn for each experimental trial. The advantages of full randomization include smaller confidence intervals on the inferred step error, the ability to use maximum likelihood analysis without heuristics, straightforward optimization of the sequence lengths, and the ability to model and measure behaviors that go beyond the typical assumption of time-independent error rates. We discuss models of time-dependent or non-Markovian errors that generalize the basic RB model of a single exponential decay of the success probability. For any of these models, we implement a concrete protocol to minimize the uncertainty of the estimated parameters given a fixed time constraint on the complete experiment, and we implement a maximum likelihood analysis. We consider several previously published experiments and determine the potential for improvements with optimized full randomization. We experimentally observe such improvements in Clifford randomized benchmarking experiments on a single trapped ion qubit at the National Institute of Standards and Technology (NIST). For an experiment with uniform lengths and intentionally repeated sequences the step error was $2.42^{+0.30}_{-0.22}\times 10^{-5}$, and for an optimized fully randomized experiment of the same total duration the step error was $2.57^{+0.07}_{-0.06}\times 10^{-5}$. We find a substantial decrease in the uncertainty of the step error as a result of optimized fully randomized benchmarking.

A boson sampler implements a restricted model of quantum computing. It is defined by the ability to sample from the distribution resulting from the interference of identical bosons propagating according to programmable, non-interacting dynamics. Here, we demonstrate a new combination of tools for implementing boson sampling using ultracold atoms in a two-dimensional, tunnel-coupled optical lattice. These tools include fast and programmable preparation of large ensembles of nearly identical bosonic atoms ($99.5^{+0.5}_{-1.6}\;\%$ indistinguishability) by means of rearrangement with optical tweezers and high-fidelity optical cooling, propagation for variable evolution time in the lattice with low loss ($5.0(2)\;\%$, independent of evolution time), and high fidelity detection of the atom positions after their evolution (typically $99.8(1)\;\%$). With this system, we study specific instances of boson sampling involving up to $180$ atoms distributed among $\sim 1000$ sites in the lattice. Direct verification of a given boson sampling distribution is not feasible in this regime. Instead, we introduce and perform targeted tests to determine the indistinguishability of the prepared atoms, to characterize the applied family of single particle unitaries, and to observe expected bunching features due to interference for a large range of atom numbers. When extended to interacting systems, our work demonstrates the core capabilities required to directly assemble ground and excited states in simulations of various Hubbard models., Comment: 20 pages, 7 figures (main text and methods); 8 pages, 2 figures (supplemental materials)

The continuing improvement in the qualities of photon-number-resolving detectors opens new possibilities for measuring quantum states of light. In this work we consider the question of what properties of an arbitrary multimode Gaussian state are determined by a single photon-number-resolving detector that measures total photon number. We find an answer to this question in the ideal case where the exact photon-number probabilities are known. We show that the quantities determined by the total photon number distribution are the spectrum of the covariance matrix, the absolute displacement in each eigenspace of the covariance matrix, and nothing else. In the case of pure Gaussian states, the spectrum determines the squeezing parameters.

Joint Gaussian measurements of two quantum systems can be used for quantum communication between remote parties, as in teleportation or entanglement swapping protocols. Many types of physical error sources throughout a protocol can be modeled by independent Gaussian error channels acting prior to measurement. In this work we study joint Gaussian measurements on two modes $\mathsf{A}$ and $\mathsf{B}$ that take place after independent single-mode Gaussian error channels, for example loss with parameters $l_\mathsf{A}$ and $l_\mathsf{B}$ followed by added noise with parameters $n_\mathsf{A}$ and $n_\mathsf{B}$. We show that, for any Gaussian measurement, if $l_\mathsf{A} + l_\mathsf{B} + n_\mathsf{A} + n_\mathsf{B} \geq 1$ then the effective total measurement is separable and unsuitable for teleportation or entanglement swapping of arbitrary input states. If this inequality is not satisfied then there exists a Gaussian measurement that remains inseparable. We extend the results and determine the set of pairs of single-mode Gaussian error channels that render all Gaussian measurements separable.

For many quantum systems intended for information processing, one detects the logical state of a qubit by integrating a continuously observed quantity over time. For example, ion and atom qubits are typically measured by driving a cycling transition and counting the number of photons observed from the resulting fluorescence. Instead of recording only the total observed count in a fixed time interval, one can observe the photon arrival times and get a state detection advantage by using the temporal structure in a model such as a Hidden Markov Model. We study what further advantage may be achieved by applying pulses to adaptively transform the state during the observation. We give a three-state example where adaptively chosen transformations yield a clear advantage, and we compare performances on an ion example, where we see improvements in some regimes. We provide a software package that can be used for exploration of temporally resolved strategies with and without adaptively chosen transformations., Comment: Submitted for publication in Quantum Science and Technology. 26 pages, 8 figures. Corrected typos in appendix, updated acknowledgements

We propose and demonstrate a protocol for high-fidelity indirect readout of trapped ion hyperfine qubits, where the state of a $^9\text{Be}^+$ qubit ion is mapped to a $^{25}\text{Mg}^+$ readout ion using laser-driven Raman transitions. By partitioning the $^9\text{Be}^+$ ground state hyperfine manifold into two subspaces representing the two qubit states and choosing appropriate laser parameters, the protocol can be made robust to spontaneous photon scattering errors on the Raman transitions, enabling repetition for increased readout fidelity. We demonstrate combined readout and back-action errors for the two subspaces of $1.2^{+1.1}_{-0.6} \times 10^{-4}$ and $0^{+1.9}_{-0} \times 10^{-5}$ with 68% confidence while avoiding decoherence of spectator qubits due to stray resonant light that is inherent to direct fluorescence detection., Comment: 7 + 6 pages, 3 + 1 figures

We present a method to estimate the amount of squeezing and temperature of a single-mode Gaussian harmonic oscillator state based on the weighted least squares estimator applied to measured Fock state populations. Squeezing and temperature, or equivalently the quadrature variances, are essential parameters states used in various quantum communication and sensing protocols. They are often measured with homodyne-style detection, which requires a phase reference such as a local oscillator. Our method allows estimation without a phase reference, by using for example a photon-number-resolving detector. To evaluate the performance of our estimator, we simulated experiments with different values of squeezing and temperature. From 10,000 Fock measurement events we produced estimates for states whose fidelities to the true state are greater than 99.99% for small squeezing (r < 1.0), and for high squeezing (r = 2.5) we obtain fidelities greater than 99.9%. We also report confidence intervals and their coverage probabilities., Comment: 8 pages, 4 figures